wjgMBWMMlS 
26 
ARCHIMEDES 
private hands. In 1491 it belonged to Georgius Yalla, who 
translated from it the portions published in his posthumous 
work l)e expetendis et fugiendis rebus (1501), and intended to 
publish the whole of Archimedes with Eutocius’s commen 
taries. On Valla’s death in 1500 it was bought by Albertus 
Pius, Prince of Carpi, passing in 1530 to his nephew, Rodolphus 
Pius, in whose possession it remained till 1544. At some 
time between 1544 and 15 64 it disappeared, leaving no 
trace. 
The greater part of A was translated into Latin in 1269 
by William of Moerbeke at the Papal Court at Viterbo. This 
translation, in William’s own hand, exists at Rome (Cod. 
Ottobon. lat, 1850, Heiberg’s B), and is one of our prime 
sources, for, although the translation was hastily done and 
the translator sometimes misunderstood the Greek, he followed 
its wording so closely that his version is, for purposes of 
collation, as good as a Greek manuscript. William used also, 
for his translation, another manuscript from the same library 
which contained works not included in A. This manuscript 
was a collection of works on mechanics and optics; William 
translated from it the two Books On Floating Bodies, and it 
also contained the Plane Equilibriums and the Quadrature 
of the Parabola, for which books William used both manu 
scripts. 
The four most important extant Greek manuscripts (except 
C, the Constantinople manuscript discovered in 1906) were 
copied from A. The earliest is E, the Venice manuscript 
(Marcianus 305), which was written betw r een the years 1449 
and 1472. The next is D, the Florence manuscript (Laurent. 
XXVIII. 4), which was copied in 1491 for Angelo Poliziano, 
permission having been obtained with some difficulty in con 
sequence of the jealousy with which Valla guarded his treasure. 
The other two are G (Paris. 2360) copied from A after it had 
passed to Albertus Pius, and H (Paris. 2361) copied in 1544 
by Christopherus Auverus for Georges d’Armagnac, Bishop 
of Rodez. These four manuscripts, with the translation of 
William of Moerbeke (B), enable the readings of A to be 
inferred. 
A Latin translation was made at the instance of Pope 
Nicholas V about the year 1450 by Jacobus Cremonensis. 
It was made 
Nicholas the 
Regiomontai 
and revised 
the Greek 
to Cardinal 
seems to ha\ 
The editu 
gium) by T1 
text was b 
app. 12) wh 
but with in 
tion was Re 
(Norimb. Ce 
A translai 
1558 contaii 
Quadrature 
the Sand-re( 
edition, but 
manuscripts 
Torelli’s e 
princeps in 
was brought 
who also c< 
and H. Th< 
edition was 
The secon 
Archimedes 
apparatus c 
course, supe 
It naturally 
chion, and g 
Floating Be 
covered Cor 
Our desc 
may suitabl 
1 The Work. 
writer in 1897