Chapter 48, vol. II, p. 259.

XXX GERBERT AND ARABIC ASTROLOGY 715
Mathematica Alhandrei in a tenth century manuscript.^ But no author is mentioned, and instead of Mathematica the title reads "Incipiunt proportiones cppfcfntfs knkstrprx in- dxstrkb," which may be deciphered as "Incipiunt propor- tiones competentes in astrorum industria." ^ Possibly there- fore this treatise is a part of the work of Alchander, and the title Mathematica Alexandri is an error for Mathematica Alhandrei.
Moreover, in later manuscripts we encounter authors Alkandri- with names very similar to Alchandrus and works by them Alchan-
of the same sort as that we have just considered. In a fif- drmus on
/-\ r 1 r ^ mi nativities
teenth century manuscript at Oxford we find ascribed to according
Alkandrinus an account of the types of men born in each ^a^sfons of the twenty-eight mansions of the moon ^ such as we have of the seen formed a part of the Mathematica Alhandrei. And in a fifteenth century manuscript at Paris occurs under the name of Alchandrinus what seems to be a Christian revi- sion of that same part of the Mathematica Alhandrei.'^ What appears to be another revision and working over of this same discussion of nativities according to the twenty- eight mansions of the moon ^ appeared in print a number of times in the sixteenth and seventeenth centuries and in
* BN 17868, f ol. I7r. The 183V ". . . finem f ecimus. _ Com- Incipit is the same as in Ash- pleta fuit hec compilatio in con- mole 369. The work here seems versione sancti pauli apostoli to be incomplete, since after fol. anno domini 1350 (1305?) vacante 17V most of the remaining leaves sede per mortem Benedicti un- of the MS (which has 21 fols. in decimi cuius anima requiescat in all) are blank. pace. Amen." It would there-
*The vowels being represented fore seem that some compiler has
by the consonants following, a made an extract from Alchandrus
common medieval cipher. on the twenty-eight mansions.
*A11 Souls 81, 15th century, * BN 10271, fols. 9r-52v, "In-
fols. I45v-i64r. "Cum sint 28 cipit liber alchandrini philosophi
mansiones lune. . . ." Coxe was de nativitatibus hominum secun-
mistaken in thinking that the dum compositionem duodecim
work of Alkandrinus continued to signorum celi, quem reformavit
fol. 188 and was in two parts, for quidem philosophus cristianus
at fol. i63r we read, "Expliciunt prout patet, quia in quibusdam
iudicia libri Alkandrini que sunt differt iste liber ab antique
in divisione triplici 12 signorum primordiali. Primo facies arietis
que sunt apparencie per certa in homine sive in masculo. Alna-
tempora super terram." More- liet est prima facies arietis. . . ."
over, the seven chapters on the * Steinschneider (1905), 30. planets which follow end at fol.
7i6
MAGIC AND EXPERIMENTAL SCIENCE chap.
Albandi- nus.
Geomancy of Alkar- dianus or Alchandi- andus.
French and English translations as well as Latin. The author's name in these printed editions is usually given as Arcandam, but the English edition of 1626 adds "or Al- chandrin." ^
Two other manuscripts at Paris ^ contain under the name of Albandinus a "book of similitudes of the sons of Adam, fortunate and unfortunate, of life or death, according to nations, that is, their nativities according to the twelve signs." The treatise opens with a method of calculating a person's nativity from the letters in his own and his mother's name similar to that which occurs in the course of the Mathematica Alliandrei, but then applies it directly to the twelve signs rather than to the twenty-eight mansions of the moon. It also does not bother with the Hebrew alpha- bet but gives numerical equivalents directly for the Latin letters. Some treatise by Albandinus on sickness and the signs in a manuscript at Munich ^ is perhaps identical with the foregoing.
To an Alkardianus or Alchandiandus is ascribed a geo- mancy,* and since it also is arranged according to the twenty- eight divisions of the zodiac with 28 judges and 28 chapters each consisting of 28 lines in answer to as many ques- tions, it would seem almost certain that it is by the same author who treated of the influences of the 28 houses or
* The editio princeps seems to be "Arcandam doctor peritissimus ac non vulgaris astrologus, de veritatibus et praedictionibus astrologiae et praecipue nativita- tum seu fatalis dispositionis vel diei cuiuscunque nati, nuper per Magistrum Richardum Roussat, canonicum Lingoniensem, artium et medicinae professorem, de con- fuse ac indistincto stilo non minus quam e tenebris in lucem aeditus, re cognitus, ac innumeris (ut pote passim) erratis expurgatus, ita ut per multa maxime necessaria et utilissima adiecerit atque adnota- verit modo eiusdem dexteritate praelo primo donatus." Paris, 1542.
The British Museum also con-
tains another Latin edition of Paris, 1553 ; French editions of Rouen, 1584 and 1587, Lyons 1625 ; and English versions printed at London, 1626 (translated from the French), 1630, 1637, and 1670.
'BN 7349, 15th century, fol. 56r, seems only a fragment of the work; BN 7351, 14th century, takes up the various signs.
*CLM 527, I3-I4th century, fols. 36-42, de physica signorum et supernascentium et aegrotantium.
*Addit. 15236, English hand of I3-I4th century, fols. i30-52r, "libellus Alchandiandi." BN 7486, 14th century, "Incipit liber alkar- diani phylosophi. Cum omne quod experitur sit experiendum propter se vel propter aliud. . . ."
XXX GERBERT AND ARABIC ASTROLOGY 717
fades of the twelve signs upon those born under them. Moreover, this Alkardianus or Alchandiandus states in his preface that he has composed certain books on the disposi- tions of the signs and the courses of the planets and on prediction of the future from them. "But since modems always rejoice in brevity," he has added this handy and rapid geomantic means of answering questions and ascer- taining the decrees of the stars. The 28 tables of 28 lines each of this Alkardianus or Alchandiandus are identical with one of the two such sets ^ commonly included in the Experimentarins ^ of Bernard Silvester, a work of geomancy which he is said to have translated from the Arabic.^ He lived in the twelfth century and will be the subject of one of our later chapters.
It still remains to speak of a portion of our tenth cen- Ananony- tury manuscript at Paris which begins, after the book of Jfse^or^^*' Alchandrus seems to have concluded, with the words, fragment "Quicunque nosse desiderat legem astrorum ..."■* This tenth Incipit is so similar to that of the twenty-one chapters on century, the astrolabe, "Quicumque astronomiam peritiam discip- linae . . ." and to that of the four books of astronomy, "Quicumque mundane spere rationem et astrorum," that one is tempted to imply some relation between them, and, in view of the tenth century date of the one at present in question, to connect it like the others with the name of Gerbert. Our present treatise or fragment of a treatise is largely astro- logical in character, "following for the present the wisdom of the mathematici who think that mundane affairs are car- ried on under the rule of the constellations." This refusal to accept personal responsibility for astrological doctrine is similar to the attitude of the author of the four books of
* The set in which the first line of life and death at f ol. I3r-v reads. "Tuum indumentum dura- "Incipit epistola Phetosiri de bit tempore longo." sphaera" separates this treatise or
' Very probably this title was fragment from the preceding
derived from the Incipit just given liber Alchandri philosophi. Also
in note 4, p. 716. this treatise is in a different and
* See Sloane 2472, 3554, 3857. slightly older hand than fols. 2- *BN 17868, fol. I4r-i6v. The 13 are, or at least such was Bub- letter of Petosiris on the sphere nov's opinion (1899), 125, note.
7i8 MAGIC AND EXPERIMENTAL SCIENCE chap, xxx
astronomy, so that perhaps the present text is the missing fragment required to fulfil his promise to treat of the sub- ject of prognostication in later chapters. If so it indulges in some repetition, as it goes into the relations existing be- tween signs, planets, and elements, and gives the "Saracen" names ^ for the twenty-eight mansions of the moon. It in- cludes a way to detect theft for each planet and a method of determining if a patient will recover by computation of the numerical value of the letters in his name. These features are suggestive of the Mathematica of Alchandrus.
* BN 17686, f ol. 14V, "que sarraceni nuncupant ita."