Skip to main content

Full text of "The Genealogy of the Editions of the "Genealogia Deorum""

See other formats


Early Journal Content on JSTOR, Free to Anyone in the World 

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in 
the world byJSTOR. 

Known as the Early Journal Content, this set of works include research articles, news, letters, and other 
writings published in more than 200 of the oldest leading academic journals. The works date from the 
mid-seventeenth to the early twentieth centuries. 

We encourage people to read and share the Early Journal Content openly and to tell others that this 
resource exists. People may post this content online or redistribute in any way for non-commercial 

Read more about Early Journal Content at 
journal-content . 

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people 
discover, use, and build upon a wide range of content through a powerful research and teaching 
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit 
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please 

Modern Philology 

Volume XVII December TQIQ Number 8 


Hortis, in his generally excellent description of the several 
editions of the Genealogia deorum of Boccaccio, does not define the 
relationships of those editions to each other, and is not always 
accurate in occasional statements bearing on those relationships.' 
The following notes will serve to define those relationships, and will 
touch on some other matters of bibliographical interest. 


Certain facts with regard to the MSS are first to be mentioned. 

In 1371 Boccaccio allowed a friend to make a copy of an auto- 
graph MS, now lost, of the Genealogia, and from that first apograph 
other copies were made. The text of the lost autograph is now 
called the Vulgate text. A second autograph, containing a revised 
text, is preserved in the Laurentian Library.^ 

1 A. Hortis, Studj sulle opere latine del Boccaccio, Trieste, 1879, pp. 769—85. I have 
examined also the descriptions by A. Bacchi della Lega in liis Bibliografia boccaccesca, 
Bologna, 1875, and those by several of the general bibliographers, but find no statements 
as to the relationships of the editions other than those by Cl§ment and Prince d'Essling 
quoted below, pp. 75, n. 1, and 78. n. 2. 

2 O. Hecker, Boccaccio-Funde, Brimswick, 1902, pp. 93 3.; H. Hauvette, Boccace, 
Paris, 1914, pp. 414, 446-48. Hecker prints from the second autograph the Dedicatory 
Letter (but not the single chapter of the general Proem, nor the Proem of Book I), the 
Proems of Books II-XIII, and Books XIV and XV entire. 

425] 65 [Modern Philology, December, 1919 

66 Ernest H. Wilkins 

To the list of the apographs' is now to be added a MS, prepared 
for and owned by Coluccio Salutati, recently given to the University 
of Chicago by President F. W. Gunsaulus, of the Armour Institute 
of Technology.^ 

The Chicago MS and all apographs previously examined in this 
regard have the Vulgate text.' 

Before each of the first 13 books in the second autograph MS 
stands an elaborate genealogical tree showing the genealogy of the 
several divinities to be discussed in that book. At the top is a large 
circle in which is written the name of a divinity. From this circle 
descends a stem, which now expands into other lesser circles, now 
sends forth leaves, and now branches, which in their turn expand 
into circles and send forth leaves and lesser branches. In the center 
of each circle or leaf a name is written.^ Similar trees appear in the 
Chicago MS and in several other MSS; some MSS however have 
blank spaces where the trees should be.* 

For each tree there appears in the second autograph MS, in the 
Chicago MS, and doubtless in the MSS in general, a special rubric 
stating the scope of the genealogy illustrated by that tree. The 
first of these rubrics reads as follows in the text of the Chicago MS 
(f. 14r): 

In arbore designata desuper ponitur in culmine demogorgon uersa in 
celum radice nee solum infra descripte progeniei sed deorum omnium 
gentilium pater, et in ramis et frondibus ab eo descendentibus describuntur 
eius filii et nepotes de quibus omnibus hoc in primo libro prout signati sunt 
distincte describitur. Verum ex eis ether solus excipitur. de quo et eius 
amplissima posteritate in libris sequentibus describetur. Fuerunt ergo 
demogorgoni filii filieque nouem. Quorum primus, litigium. secundus. 
pan. tercia cloto. quarta lachesis. quinta Attropos. sextus polus seu poUux. 
Septimus phyton seu phaneta. octaua terra. Nonus autem herebus. 

The autograph MS contains 45 passages from Homer, transcribed 
in Greek letters.* The apographs vary in their treatment of this 

1 See Hortis, pp. 227, 388, 919-23, and Hecker, p. 97, n. 1. 

2 In a forthcoming monograph I shaU discuss tUs MS In detail. 
' Hecker, p. 97, n. 1. 

* Hecker, pp. 94-95, and Plate XIX. 

» Hortis, pp. 919-23. In a forthcoming monograph I shall discuss Boccaccio's 
trees in detail. 

« Hecker, pp. 137-53. 


Genealogy of the "Genealogia Deorum" 67 

material. The Chicago MS and some others reproduce it all, 
others give it in part, others omit it altogether.^ The first passage, 
in Book III, chapter 22, consists of four lines (JKad xiv. 214-17), 
and the second passage, in Book IV, chapter 18, consists of three 
lines {Iliad i. 402-4). 

In the Chicago MS, immediately after the end of the text of the 
Genealogia, stands a series of 17 hexameters headed Versus ediii per 
insignem uiruni ser Dominicum siluestri de Florentia super continentia 
lihrorum de Genealogia deorum darissimi uiri domini Johannis 
boccaccij de Certaldo? The first of the lines reads: 

Que narrat ter quinque libris boccacius audi. 

Each of the next 15 lines indicates the content of one of the fif- 
teen books of the Genealogia. The seventeenth line reads : 
Hoc ter quinque libris epigrama dominicus addit. 

In a MS of the Laurentian Library (Ixxxx Inf. 13) containing poems 
by Domenico di Silvestro and others these same hexameters appear, 
followed by an eighteenth, which reads : 

Quem genuit ripis Florentia fluminis Arni. 

A note upon the margin of the MS states that the last two lines 
were composed by Coluccio Salutati.* The authority of this marginal 
note, accepted by Hortis,* may well be doubted, in view of the fact 
that in the Chicago MS, which, as stated above, was prepared for 
and owned by Coluccio, the line Quem .... Arni does not appear, 
and there is no indication of any sort that the line Hoc .... addit 
differs in authorship from the preceding lines. 


The Genealogia was first printed in 1472 at Venice by Wendelin 
of Speier. Seven other editions appeared in the late fifteenth 
century and the early sixteenth : the second edition in 1481 at Reggio, 

1 Hortis, pp. 388, 919-23; Hecker, p. 138, n. 2. 

2 Domenico di Silvestro was a notary and writer of Latin verse, active in the period 
1364-1407. See L. Mehus, in Ambrosius Traversarius, Epistolae, Florence, 1759, 
Vol. II, pp. cccxxvi-cccxxxi. 

'Mehus, p. cccxxx; A. M. Bandini, Cat. cod. lat. bib. med. laur.. Vol. Ill, Florence, 
1776, coU. 714-15. 

< Pp. 770, 771, 777-81. The statements of Hortis with regard to the Versus are 
somewhat obscure: the statement liere given should make the matter clear. 


68 Ernest H. Wilkins 

the third in 1487 at Vicenza, the fourth in 1494 at Venice, the fifth 
in 1497 at Venice, the sixth in 1511 at Paris, the seventh in the same 
year at Venice, the eighth in 1532 at Basle. There is no more recent 
print of the work as a whole.^ 

The text of each of these editions is the Vulgate text.^ 

VENICE, 1472 

This edition contains, first, the Table of Rubrics; second, the 
Genealogia itself; third, the Alphabetical Index by Domenico 
Bandini; fourth, the Versus of Domenico di Silvestro. 

The printer did not undertake to reproduce the genealogical 
trees which stood presumably in the MS which served him as copy. 
At the point (f. 15r) where the first tree should have appeared, he 
introduced a heading consisting of the first tree rubric with the 
substitution of the words: "Hie secundum exemplar deberet esse 
arbor signata in qua" for the first three words of the rubric. He 
then left blank a space of half a page, so that the tree might be filled 
in by hand. Similarly he left spaces, varying in size from two- 
thirds of a page to an entire page, for the other twelve trees. For 
the second and later books he entered no heading for the trees, and 
omitted the tree rubrics entirely from the text. 

For the first Greek quotation the edition of 1472 prints the first 
two lines only, without division between lines or between words, thus : 
For the second quotation the first few letters only are printed, thus : 
aJxeKaroTx. with the words et caetera immediately following. For 
the remaining passages the Greek letters are omitted entirely, in 
this edition and in all later editions.^ 

The Versus of Domenico di Silvestro end with the seventeenth 
line. Hoc .... addit, and there is no indication of a different 

1 1 have examined the several editions in the following copies: 1st, University of 
Chicago and Harvard; 2d, Newberry, Annmary Brown Memorial, and Library of Con- 
gress; 3d, Library of Congress; 4th, University of Chicago and Harvard; 5th, Library 
of Congress; 6th, University of Chicago and Harvard; 7th, Harvard; 8th, Harvard. 
Hortis, pp. 769-70, mentions also a compendium printed at some time before 1500. 
For Hecker's partial print from the second autograph, see above, p. 65, n. 2. 

2 Hecker examined the editions of 1497, Paris 1511, and 1532 in this regard: op. cit., 
p. 97, n. 1. I have examined the other five editions. 

2 On the printing of Greek passages in Latin books issued before 1476, see R. Proctor, 
The Printing of Greek in the Fifteenth Century ( = Bibliographical Society, Illustrated 
Monographs, No. VIII), Oxford, 1900, pp. 24-48. 


Genealogy of the "Genealogia Deorum" 69 

authorship for that Hne. These statements apply also to the print- 
ing of the Versus in all later editions in which they are contained. 

The first 12 quires in this edition are of 10 leaves each; and in 
these quires each full page has 41 printed lines. The thirteenth 
quire has, however, 12 leaves (ff . 121-32) ; and in this quire each full 
page has 42 printed lines. Quires 14-18 have 10 leaves each; quire 
19 has 6 leaves; quires 20-22 have 10 leaves each; quires 23-25 have 
8 leaves each ; and quires 26 and 27 have 10 leaves each. In quires 
14-27 each full page has 41 printed lines. Both sides of f. 242, the 
last leaf of quire 25, are blank. This leaf stands between chapters 
18 and 19 of Book XIV. There is no textual reason for the leaving 
of space at this point: it is simply an instance of bad planning in the 
assignment of material to quires. The text of the Genealogia ends 
on f. 259r. The Index begins on f. 259v. The colophon appears 
on f. 295v.i 

Hortis notes two varieties of this edition, the second variety 
being more correct in text in the first 50 pages.* 

In the next year, 1473, Wendelin of Speier printed an edition of 
the De Montibus uniform with the 1472 edition of the Genealogia* 

KEGGIO, 1481 

This edition and all later editions contain both the Genealogia 
and the De Montibus. 

This edition, in the part containing the Genealogia, agrees with 
the edition of 1472 in contents, in the fact and size of the spaces left 
for the trees, in the retention of the first tree rubric, as modified, 
and in the omission of the other tree rubrics from the text. It agrees 
in the treatment of the Greek quotations, except that the accents and 
breathings are omitted, and that five additional errors appear in 
the first passage: (Tr]de<T(j>iv for (TTri9e<T(t>iv; ekvaaro for eXucraro; Kevov 
for Kearov; deKr]Tripai for deKrp-ijpia; and iraijr for iravT. 

The first 12 quires are of 10 leaves each; and in these quires each 
full page has 41 printed lines. Comparison of this portion of the 

' This edition has no signatures. My statements as to the number of leaves in the 
several quires are taken from Marie Pellechet, Catalogue giTiiral des incunabulea des 
biblioihkques publiques de France, Vol. II, Paris, 1905, p. 30, No. 2466. 

2 The Chicago copy is of the first variety. The Harvard copy is of the second 

> Hortis, p. 774, No. XV. 


70 Ernest H. Wilkins 

edition with the corresponding portion of the edition of 1472 shows 
that each page in the 1481 edition bears exactly the same material 
as the corresponding page in the 1472 edition. 

The thirteenth quire in the 1481 edition has 14 leaves (ff . 121-34) ; 
but in this quire the distribution of the material is such that the 
quire as a whole contains exactly the same material as the corre- 
sponding 12-leaf quire in the 1472 edition. The numbers of the 
printed lines on the several pages of this quire are respectively: 
41, 41, 42, 42, 39, 39, 37, 37, 41, 39, 41, 39, 41, 41, 40, 41, 41, 41, 
35, 41, 40, 35, 36, 40, 42, 42, 40, 0. Spaces are left between chapters 
in such a way as to bring the total page length to the equivalent of 
40 or 41 lines. F. 121r corresponds exactly to f. 121r of the 1472 
edition, but f. 121v contains a few less words than f. 121v of the 
1472 edition; and from that point on the disparity increases up to 
f. 133v, which ends with the same word as f. 13 Iv of the 1472 
edition. F. 134r bears the same material as f. 132r of the earlier 
edition; and f. 134v is blank (being reserved for the tree of Book 
VIII), as is f. 132v of the earlier edition. 

Quires 14-18 have 10 leaves each ; quire 19 has 6 leaves; quires 
20-22 have 10 leaves each; quires 23-25 have 8 leaves each;i and 
quire 26 has 10 leaves. In quire composition, therefore, the edition 
of 1481 follows that of 1472 exactly from quire 14 through quire 

26. In all these quires each full page has 41 printed lines. Each 
page in this portion of the edition bears exactly the same material 
as the page of the 1472 edition, whose number is less by two. As in 
the 1472 edition, the last leaf of quire 25 (f. 244 in the 1481 edition) 
is left blank^ — a striking illustration of the mechanical nature of the 

Quire 27 has but 6 leaves (ff . 255-60) ; but in this quire the 
distribution of material is such that the quire as a whole contains 
exactly the same material as the first 13 pages (ff. 253r-59r) of quire 
27 of the 1472 edition. The number of printed lines on each full 
page remains 41 throughout the quire. The saving in space is made 
by crowding the composition. 

The text of the Genealogia ends on f. 260v, the last page of quire 

27. F. 261r is blank. The Index begins on f. 261v. The colophon 

1 Hortis wrongly assigns 7 leaves to quire 23, being misled doubtless by the fact 
that the last leaf of the quire, f . 244, is blank. 


Genealogy of the "Genealogia Deorum" 71 

appears on f . 297v. Each page of the Index bears exactly the same 
material as the page of the 1472 edition, whose number is less by two. 

The text in this edition and in all later editions agrees with that 
of the second variety of the 1472 edition in the points in which the 
two varieties of that edition differ. 

The portion of the edition of 1481 containing the De Montibus is 
separately paged. The pagination is identical with that of the 
edition of 1473, as inspection of the figures in the descriptions by 
Hortis will at once show. The De Montibus in the 1481 edition, 
however, lacks the last two sections: De stagnis et paludibus and 
De nominibus maris. 

It is then evident that the edition of 1481 is a page-for-page copy 
of the 1472 edition of the Genealogia (except in quires 13 and 27), 
and of an incomplete copy of the 1473 edition of the De Montibus.^ 

vicenza, 1487 

This edition is printed in two columns, and is the only one of the 
editions to be so printed. 

In the part containing the Genealogia this edition agrees with the 
earlier editions in contents, in the omission of the trees, in the reten- 
tion of the first tree rubric as modified, and in the omission of the 
other tree rubrics from the text. It leaves small spaces for the first 
five trees, no spaces for the other eight. It does not attempt the 
Greek quotations, but leaves spaces for the first two. 

The portion of the De Montibus contained in this edition ends 
at the same point as the portion contained in the edition of 1481. 

It is then evident that the edition of 1487 is derived from that 
of 1481.2 

VENICE, 1494 

This edition, in the part containing the Genealogia, agrees with 
the earlier editions in contents. 

1 Hortis, p. 777, says: " Questa edizione, tuttochfi migllore quanto alia punteggiatura, 
6 quanto alia lezione men corretta della Vindelinlana, e fu cagione dl parecchl errorl, 
accolti poi dalle edizioni successive. Per le citazioni del greco 6 inleriore alia stampa 
del 1472." The genealogy of the editions, as established by this paper, will show that 
the errors of this edition can hardly have affected any later edition except that of 1487. 
Hortis' statement as to the Greek passages is a careless generalization. He seems in 
general to have inspected only the first Greek passage in the several editions. 

« Hortis, p. 778, says: "Questa edizione segue il testo dclla Reggiana." 


72 Ernest H. Wilkins 

In this edition the genealogical trees are finally achieved, and 
that fact is announced in the title: Genealogiae loannis Boccatii: 
cum demonstrationibus in formis arborum designatis. The trees 
correspond in general to those of the second autograph and the 
Chicago MS. It is therefore evident that the designer based his 
trees upon those in a MS or MSS of the Genealogia, or upon trees 
entered by hand in a printed copy which were themselves derived 
from a MS source. Two main differences appear, however: the 
initial circle in each case contains a portrait or symbolic image of 
the divinity in question; and all the trees except those for Books 
I, III, and VIII bear scrolls instead of leaves. For purposes of 
comparison with later editions it may be noted that the portrait 
of Demogorgon in Tree I shows a rather old man with a clearly 
defined crown; that in Tree IX the scroll for Hebe springs alone 
from the main stem; that in Tree XII the head of the dart is not 
inked in; and that in Tree XIII the dart is held in the right hand of 
Jupiter. The names in all trees are printed in Gothic letters. Over 
each tree is placed that one of the Versus of Domenico di Silvestro 
which refers to the book in question. The first tree rubric no longer 
appears in the text. 

In this edition and in the next three editions the perfection and 
the regular alignment of the letters in the names printed in the 
leaves and scrolls indicate that these names were printed from type 
set in cavities in the blocks. 

The treatment of the Greek quotations is like that of the first 
edition, except that the accents and breathings are omitted, and 
that two additional errors occur; crr]de(T4>i,v for (TTy]di<j(jiiv in the first 
passage, and exaT-o^x for ^naToyx in the second passage. 

The pagination is quite different from that of the earlier editions. 

The De Montibus stands complete in this edition. 

It is then evident that the edition of 1494 is derived from the 
1472 edition of Genealogia and the 1473 edition of the De Montibus. 

VENICE, 1497 

This edition agrees with that of 1494 in title, in contents, and 
in the treatment and errors of the Greek passages. The trees are 
very evidently imitated, tree for tree, from those of 1494. Tree I, 


Genealogy of the "Genealogia Deorum" 73 

however, bears scrolls instead of leaves, and the names in Trees I, III, 
and VIII are printed in Roman letters. Minor variations in drawing 
appear : for instance, the portrait of Demogorgon in Tree I shows a 
rather young man, with a scarcely visible crown. 

In pagination this edition agrees exactly with that of 1494. This 
fact is suggested by inspection of the figures in the descriptions by 
Hortis, and may be confirmed by comparison of any two correspond- 
ingly numbered leaves. 

It is then evident that the edition of 1497 is a page-for-page copy 
of that of 1494.1 

PARIS, 1511 (August) 

This edition adopts a more elaborate title: Genealogie Johannis 
Boccacij cum micantissimis arborum effigiacionibus cuiusque gentilis 
dei progeniem, non tarn aperte quam summatim declarantibus Cumque 
praefoecunda omnium quae in hoc libro sunt ad finem tabula. 

In contents it differs from its predecessors by the presence of 
some editorial material and by the omission of the Table of Rubrics 
and of the Versus of Domenico di Silvestro after the Alphabetical 

The editor, loannes Kierherus, in a letter printed on the verso of 
f. 1, promises a collation of editions: 

Itaque velim, Humanissime Gotfride sic tibi persuadeas: me diligen- 
tissime coUatis exemplaribus : earn operam daturum: vt boccacius, si non 
omnibus (quis enim omnia Argi more peruideat) tamen plurimis mendis 
vindicatus: quam emendatissime in publicum prodeat. 

In a final note ad ledorem, just before the colophon, he asserts that 
the true readings have been restored in many cases. 

The trees are virtually the same as in the two preceding editions. 
In respects in which those editions differ, the Paris edition agrees 
with that of 1494: Tree I bears leaves, not scrolls, and the names in 
Trees I, III, and VIII are in Gothic letters; the portrait of Demo- 
gorgon shows a rather old man with a clearly defined crown. New 
variations in drawing appear: for instance, in Tree IX the scroll 
for Hebe is detached from the main stem and joins the scroll for 
Hyperulus, and in Tree XII the head of the dart is printed in solid 

' Hortis, p. 780, says: " Questa edizione non segue U testo della Reggiana." 


74 Ernest H. Wilkins 

The treatment and errors of the first Greek passage are exactly 
the same as in the two preceding editions. The second Greek passage 
is omitted. 

F. Ir bears the title, and f. Iv editorial material. The Proe- 
mium begins on the recto of what is actually the second leaf (the Table 
of Rubrics being omitted, as stated above). This second leaf, how- 
ever, is numbered VI, as in the two preceding editions, and all later 
leaves are numbered accordingly. From the second leaf on, this 
edition agrees exactly in pagination with the two preceding editions. 
This fact is suggested by inspection of the figures in the descriptions 
by Hortis, and may be confirmed by comparison of correspondingly 
numbered leaves. 

It is then evident that the edition of Paris, 1511, is a page-for- 
page copy, save for the minor variation in contents, of that of 

The omission of the Versus of Doinenico di Silvestro must have 
been deliberate. The leaf of the 1494 edition which bears the 
Versus must have been complete in the copy used by the Paris printer, 
since the material on the verso of that leaf is completely represented. 
The omission was perhaps due to a feeling that since most of the lines 
appeared earlier in the edition as superscriptions for the trees a 
repetition at this point would be undesirable. 

VENICE, 1511 (November) 

This edition agrees with those of 1494 and 1497 in title, in 
contents, and in the treatment and errors of the Greek passages, 
except that the second word of the first passage becomes a/ct instead 

of KM. 

The trees are virtually the same as in the three preceding editions. 
In respects in which the editions of 1494 and 1497 differ, the edition 
of Venice, 1511, agrees with that of 1497: Tree I bears scrolls, and 
the names in Trees I, III, and VIII are printed in Roman letters. 
The new variations of the Paris edition are not represented in that 
of Venice, 1511. 

• Hortis, p. 782, wrongly says of this edition: "Con alberi genealogici proprl." He 
notes, p. 783, the peculiar numeration of the second leaf, but evidently does not under- 
stand its cause: "Nell' esemplare di mia proprleti, e negli altrl veduti da me, il foglio 
secondo porta erroneamente il numero VI, laddove la segnatura 6 esattamente a. ij." 


Genealogy of the "Genealogia Deokxjm" 75 

The colophon claims a revision of the text: 

Habes lector peritissime: loannis Boccatii deorum genealogiam nouiter 
multis erroribus expurgatam: et in pristinum candorem deductam. 

In pagination this edition agrees exactly with those of 1494 and 
1497. This fact is suggested by inspection of the figures in the 
descriptions given by Hortis, and may be confirmed by comparison 
of correspondingly numbered leaves. 

It is then evident that the edition of Venice, 1511, is a page-for- 
page copy of that of 1497.^ 

BASLE, 1532 

This edition, edited by Jacob Micyllus, differs in many respects 
from its predecessors. The title is quite different. The volume 
contains, first, editorial material; second, a new Alphabetical Index 
prepared presumably by Micyllus himself; third, the Table of 
Rubrics; fourth, the Genealogia; fifth, the De Montibus. The 
Versus of Domenico di Silvestro do not appear after the Genealogia. 

In the introductory editorial letter Micyllus asserts that he has 
corrected many textual errors, states that he has rewritten some 
passages, and indicates knowledge of a single eariier edition : 

Deinde autem, quod ad nostram operam attinet, pleraque in hoc, ita 
nunc emendauimus, atque restituimus, ut si nostrum Bocatium, cum eo 
qui ab alijs ante excusus habetur comparaueris, ab innumeris, adeoque 
incredibilibus mendis ac uitijs repurgatum depraehendes. Non enim modo 
uoces hie atque illic quasdam expleuimus, aut immutauimus, id quod in 
alijs plerumque fieri solet, sed totas alicubi fabulas retexuimus, suisque ueris, 
et a doctis ac ueteribus traditis, capitibus, ac locis restitutas rescripsimus. 

Trees III, IV, VI, and IX-XIII agree exactly, in every detail 
of the design, with the corresponding trees of the Paris edition. 
When corresponding trees are placed and studied side by side, it is 
evident beyond a possibility of doubt that the printer of the Basle 
edition used for these trees the actual blocks used by the Paris 

> Hortis lists and treats the edition of Venice, 1511. publislied in November, before 
that of Paris, 1511, published in August. He says, p. 781, "Con alberi genealogici 
dlfferenti da quelli dell' edizione veneta di O. Seoto (1494), uguali a quellidell' edizione 
veneta di Manfredo da Streuo (1497)." Both the differenli and the uguali imply too 
much. Hortis quotes C. Clfiment, Bibliolheque Curieuse, Vol. IV, Hanover, 1753, p. 331, 
n. 69, as saying of this edition: "elle parolt avoir StS faite sur celle de 1497, dont elle a 
conserve les fautes." Prince d'Essling, les Livres d figures vcnitie.ns de la fin du XV 
sikcle et du commencement du XVI^ , Part I, Vol. II, Florence, 1908, p. 240, says of this 
edition: "12 arbres genealogiques, mauvaises copies des bois de I'edition 1494." 


76 Ernest H. Wilkins 

printer. The only differences of any sort that I can detect are a 
few corrections in the spelling of names, and these very differences 
confirm the recognition of the fact that the same blocks were used, 
for the corrections are made in nondescript letters, evidently cut for 
this special purpose, and are made in such a way that as little as 
possible of the old name has to be cut out. Tree IV, for instance, 
has six corrections. Whereas the Paris edition had 

Borream, Oetam, Purpureum, Piridilem, Chroniam, Phytonem, 
the Basle edition has 

Boream, Aetam, FnTphijrionem, Proydilem, Coronidem, Fhileto, 

the letters which I have italicized being the only ones that are 
newly cut. I infer that the type used by the Paris printer for the 
tree names had been so fastened as to become virtually an integral 
part of the blocks. 

The other five trees are, however, of a new and simpler sort. 
Circles replace leaves and scrolls throughout, and names instead of 
portraits or images appear in the initial circles. I infer that the 
Basle printer was unable, perhaps because of loss, to secure the Paris 
blocks for these five trees. 

For the first Greek quotation Micyllus gives in the text the full 
four lines. For the second and later quotations he enters no Greek 
in the text. In the second and in most, but not all, later cases he 
supplies the Greek passages, wholly or in part, in the notes appended 
to the several chapters. 

That Micyllus in preparing the text of his edition of the Genealogia 
used only a single earlier edition, as indicated above, and that he 
did not use a MS, is made evident by the following passage in the 
introductory editorial letter which appears in his edition of Hyginus:' 

Proinde cum superiore anno genealogias deorum, perinde ut a Bocatio 
ante annos aliquot coUectae fuerant, hortatu amici nostri Joannis Heruagii 
relegissem et easdem ab innumeris mendis, quibus aeditio prior inuoluta 
atque obruta erat, castigatas excudendas dedissem: atque idem nuper alium 
quendam uetustum ac manu scriptum codicem, in quo per capita easdem res, 
atque idem argumentum ab Hygino (sic enim inscriptus liber is erat) trac- 
tabitur, uisendum et quoad eius fieri potest, emendandum quoque et 
restituendum dedisset, ego .... uolui laborem hunc meum excellentiae 
tuae .... dedicare. 

> Basle, 1535. 


Genealogy of the "Genealogia Deorum" 77 

The use of the Paris blocks for eight of the trees, the omission of 
the second Greek passage from the text, and the omission of the 
Versus of Domenico di Silvestro make it evident that the Basle 
edition was based on that of Paris. The presence of the Table of 
Rubrics shows that some other edition was used to supply that 
particular material.' 


We have then the following stemma for the editions, a heavy line 
indicating a page-f or-page copy : 

Venice, 1472 

Reggie, 1481 Venice, 1494 

Vicenza, 1487 Venice, 1497 Paris, 1611 

Venice, 1511 Basle, 1532 

The four Venetian editions, it will be seen, constitute a direct 
line of descent. 

From the foregoing material it is evident that all editions of the 
Genealogia are derived directly or indirectly from that of 1472. My 
examination of the editions shows no indication that any editor of 
a later edition had recourse to a MS, except that the editor of 
the edition of 1494 may have derived his genealogical trees from a 
MS source. The mechanical dependence of each edition upon a 
predecessor indicates in general that MSS were not used. Micyllus 

I Hortis, p. 389, n., says that this edition "ommette pift volte 1 caratterl grecl." It 
omits them, as indicated above, in every case except the first. 


78 Eknest H. Wilkins 

certainly did not use a MS. I see no indication that the textual 
emendations announced in the last three editions are anything more 
than conjectural. 

The edition of 1472 is then the best printed representative of the 
Vulgate text of the Genealogia, and should be cited, in preference to 
the edition of 1532, for all portions of the Genealogia, except those 
printed by Hecker from the autograph,* and for any citation in 
which the reading of the Vulgate text as against that of the auto- 
graph is desired.^ 

Ernest H. Wilkins 
UNrvERSiTY OP Chicago 

> See above, p. 65, n. 2. 

' Hortls calls the edition of 1532 the best (pp. 389, n., 785). He himself notes that 
the edition of 1472 is abbaetama esatta, and that at certain points it gives a better reading 
than that of 1532 (p. 773); and that the edition of 1532 "si allontana assai spesso da" 
codici pia autorevoli" (p. 785); and he quotes (ibid.) Clfiment (Bib. Cur., Vol. IV, 
p. 333, n.) as saying: "Micyllus n'a pas toujours 6t6 heureux dans ses corrections, parce 
qu'il n'a pas conf6r6 les dlverses Editions de Bocace, qui avoient vtt le jovir avant I'an 
1531, et qu'il a donnS trop hardiment lieu & ses coniectures."