View Diagrams  Download Paper  Download Images 


This is a retrospective of the JCB atlas of 1511 which has 7 sheets of charts and an eighth a Ptolemaic Planisphere. The scales vary considerably with two charts having only the Adriatic area and the Aegean Sea covering the complete sheet, then to the whole Mediterranean Sea basin on a single sheet, and all of the sheets are identical in size at c39 x c57.6cms. To achieve that the measurement or scale bars had to be seriously manipulated by alterations to the scale as is discussed and proven.

ChAVM/3/D01 ChAVM/3/D02

ChAVM/3/D03 ChAVM/3/D04

ChAVM/3/D05 ChAVM/3/D06


It will come as no surprise to anybody that each and every map/chart/planisphere requires to be measured by known units not only to achieve a scale but to determine that the required plot will fit on a page/sheet of known dimensions. Without a known measurement system nothing can be achieved. Thus an analysis of the size of the sheets and the scales used will provide guidance to the actual construction of the atlas pages.

Seven of the pages can be evaluated as a group and the eighth, the Ptolemaic style chart will be evaluated singly. However, the 7 charts are divided into those with and without bordures and are measured accordingly, overall or to the internal bordure via their scale bars.

Thus we have;
JCB –08658-008; No bordure; 1100 x 2450 miliaria; Diagram ChAVM/3/D01
JCB-08658-002; Bordure; 3020 x 4460 miliaria; Diagram ChAVM/3/D02
Note, sheets 008 and 002 are drawn to the same scale and co-join to form one chart.
JCB-08658-004; Bordure; 1840 x 2680 miliaria; Diagram ChAVM/3/D03
Note, a Windrose comprises 92 units and thus this chart has a perfect plot.
JCB-08658-007; Bordure; 1580 x 2240 miliaria; Diagram ChAVM/3/D04
JCB-08658-003; Bordure; 1580 x 2240 miliaria; Diagram ChAVM/3/D05
Note, these are drawn to the same scale and produce a Mediterranean Sea portolan.
JCB-08658-005; No bordure; 600 x 850 miliaria; Diagram ChAVM/3/D06
JCB-08658-009; Outer bordure; 600 x 850 miliaria; Diagram ChAVM/3/D07

Note, these two charts are drawn to the same scale and could have been co-joined
except for the missing section around Corfu, which may have been caused by a setting out error, thus if the Aegean Sheet were moved easterly it would provide the space required.

The sheets are given as c39 x c57.6cms (c390 x c576mms) and had to be made to suit measurements used in Naples c1509AD, or of course, measurements that Vesconte Maggiolo has carried with him from Genoa! They have obviously been sized from a slightly larger sheet as there are no ragged edges and the corners have been rounded. Thus the final size is intended and no doubt used and known accurately by the cartographer.

Italy in the 15th and 16th centuries had a variety of measurements for the “foot”, some obviously meant to be the same but varying as time progressed and the manufacture of the scale bar ruler varied in accuracy. They were all handmade and there was no real system for ensuring their complete accuracy, however, still in common usage was the Roman Pes or foot of 11.64706 statute inches/ 295.835mm. It was subdivided into 12 uncia (inches) or 16 digitus (fingers) and they are respectively 24.6529mm or 18.4697mm. The atlas page is given as c390 x c576mm and appears to be 16 uncia which equals 394.45mm by 23.3 uncia which equals 575.15mm and given that these figures are within a measurement tolerance, (we do not know the accuracy of the makers scale bar,) they are an acceptable putative suggestion of the original measurement.

That can be shown to be a reasonable assumption by the following investigation as
Sheets 009 and 005 both measure a precise 600 x 850 miliaria via their scale bars. Thus we can calculate what the 50 miliaria sub-division of the scale bar was as a measurement.

Height; 600 miliaria = 394.45mm and is thus 600 x 0.657417mm per unit. Therefore 50 scale bar units = 32.8705mm or 1.33335 Uncia or 9 units per “foot”.

It is thus a simple 1 1/3rd uncia per 50 unit division and thus how easy it must have been for Vesconte Maggiolo to utilize and formulate the scales of the following sheets.

Those sheets have varying scale bars but are compared with 850 miliaria as a common length as shown and corresponding to the sheet 008 scale bar.
JCB-08658-008; 850 miliaria = 73mm
JCB-08658-002; 850 miliaria = 73mm (scale bar 1000 miliaria = 85.5mm)
JCB-08658-004; 850 miliaria = 118mm (scale bar 900 miliaria =125mm)
JCB-08658-007; 850 miliaria = 146mm (scale bar 550 miliaria = 94.5mm)
JCB-08658-003; 850 miliaria = 146mm ( scale bar 550 miliaria =95mm)
JCB-08658-009; 850 miliaria = 401.625mm ( scale bar 200 mil = 94mm)
JCB-08658-005; 850 miliaria = 401.625mm (scale bar 200mil = 94.5mm)

I have shown that the overall sheet is 600 x 850 miliaria and thus from those figures the 50 miliaria sub-division is 1.33335 or 1 1/3rd uncia. Thus if we commence with sheets 009 and 005 as the starting point because they are the largest scale and it is their size which matches the sheet at 600 x 850 miliaria we can evaluate the other charts as they descend in scale.

From the above table the basic lengths for 850 miliaria are;
009/005 = 401.625
007/003 = 146
004 = 118
008/002 = 73 and the corresponding reduction are as follows;
401.625 divided by 146 = 2.75
146 divided by 118 = 1.25 146 x 2.75 = 401.5 and 118 x 1.25 = 147.5
118 divided by 73 = 1.618 73 x 1.618 = 118.114.

This produces the most surprising result in that the change from sheet 004 to 008/002 is governed by the Golden Ratio of 1.6180339887 etc. This is an irrational number and determined by the following; “in mathematics two quantities are in the Golden ratio if their ratio is the same as the ratio of their sums to the larger of the two quantities”.

This was described by Leonardo of Pisa (Fibonacci 1170-1250) and is quantified by simple formulae and is illustrated in the appendix to this Section 1 text in diagrammatic form.

It was also described in the Golden Rectangle form and became the typical proportion for a printed book page in the medieval era when the “beautiful” page proportions were either 2:3 or 1:root3 and the Golden Section.

However our sheet is 16 x 23.33 or a 1:1.45625 ratio and quite close to the 1:root2 ratio of 1: 1.4142. That leads to the speculation that the original sheet size as manufactured was larger as I have already opined and reduced to suit the charts and atlas accordingly.

Thus sheets 008/002, the Portolan of the Mediterranean Sea and Africa combined are increased via their scale bar by 1.618 to produce the scale bar for the Atlantic seabord chart 004. Then sheet 004 is increased by 1.25 to produce the charts 007/003 and forms the whole Mediterranean Sea basin when co-joined. They in turn are increased by 2.75 to produce the large scale charts of the Adriatic and the Aegean as sheets 009/003.

Of course I suspect that the setting out was not an expansion but a reduction from the sheets 009/003 which fill the complete sheet as 600 x 850 miliaria.

FINAL COMMENT; if anybody thinks the figures do not appear exactly the same, remember I am scaling from a scanned copy and just by adding or subtracting a minute amount of a millimetre from each measurement they could be perfect. That however would be counter- productive as we do not know the accuracy of the original scale bars or the accuracy of the scale ruler used to draw them in the first place. It is thus my 60 years of draughtsmanship, using scales etc which leads me to these conclusions. A later section of this research will prove the inaccuracy of some scale bars as drawn.



With the knowledge that the Roman Uncia is the basis of the setting out the last of the charts can be evaluated, and is commenced with the proportions of longitude to latitude at the Equator which is a 14:9 ratio for a 10 degree section as drawn. Then measurements were taken at 60N, 30N, 0N and 30S and are as follows:
60N = 2/3rds uncia
30N = 1+1/3rd uncia
00N = 2 uncia
30S = 2+ 2/3rds uncia
and the whole is therefore based upon the spacing at the Equator of 10 degrees longitude being 2 Uncia; a very simple proportional setting out. That proportional setting out led to a simple geographical analysis of the chart (a continuation of ChAVM/1) to understand the comparative differences.

I therefore decided that the drawing of a square graticule chart based upon the Ptolemaic Planisphere would give a better understanding of what was actually being shown on JCB-08658-006, as diagrams ChAVM/3/D11 and D12 illustrate. The planisphere it produces is quite remarkable in its scope combining the latest knowledge of both the West Indies and the Northern hemisphere whilst maintaining the Ptolemaic scheme from the Persian Gulf/Caspian Sea to India.

ChAVM/3/D09 ChAVM/3/D10

ChAVM/3/D11 ChAVM/3/D12

When the BSB-Hss Cod.icon.135 atlas is examined later in this text it will be clearly seen as the original and the development of the planisphere leads to the marvellous Vesconte Maggiolo planisphere of 1531.

APPENDIX TO SECTION 1; Golden Ratio and scale reduction methodology

The diagram ChAVM/3/D13 clearly illustrates the methodology of the reduction by draughtsmanship, just plain geometry, from the largest scale bar sheets 009/005 to the smallest scale bar sheets 008/002. The change can be achieved in the simplest manner and the new scale bar appears automatically. Accuracy is everything in draughtsmanship and the methodology will certainly account for minor aberrations in the scale bars we record on certain charts. This method is preferable to the use of proportional compasses.



Introduction; Vesconte Maggiolo has been lauded by historians as a master cartographer for some decades, but his early life is clouded by lack of information. As already stated he was born c1470, but his death is correctly recorded in the Notarial entries for the Maggiolo Family. These omit the period from c1508/c1518 when Vesconte Maggiolo was in Naples and married to an unknown person.

This section of the text therefore concentrates upon the actual drawn pages and their comparison to the 1512 and 1519 atlas.


I have already opined that this atlas was drawn in 1510 as it is endorsed as follows;
“Vesconte de Maiolo civis Janue composuy in Neapoles de anno 1511, die XX January”.

The atlas has 10 sheets but only sheets JCB 08658-007; 004; 003 and then sheets 008 and 002 which have been combined are capable of being compared to the Parma Atlas of 1512 which has only 4 sheets extant.


The atlas is endorsed as follows; “Vesconte de Maiolo civis Janue composuy hanc cartam in Neapoly de anno domini 1512, die X Marcij”





Thus we can establish that this atlas was finished approximately 14 months after the 1511 atlas and if the time scale was only for this atlas it points to an atlas of at least 7 pages in total. However I draw your attention to the fact that the Portolan Chart held by the Hispanic Society of America, New York, reference K33 is endorsed as follows; “Vesconte Maiolo composuy hanc cartam in Neapoli de anno 1512, die II Junii”. That is only three months after the date of the 1512 atlas and indicates it is either a finishing date for a Portolan Chart being constructed whilst the two previous atlases were being drawn or it was carried partially complete from Genoa and thus started before 1509. Note that the 1513 Portolan Chart is dated 29th August, some 14 months later.

The four charts of the Parma 1512 Atlas are thus compared to the matching four (five) charts of the 1511 Atlas.





By drawing the two pages with the Windrose alignments parallel and aligned at the Pillars of Hercules it can be readily seen that in fact there is a slight twist in the comparative Windrose setting out. If the 1512 chart was twisted anti-clockwise there would be a greater alignment with the scale bars equalised as the second diagram illustrates. There is now greater concordance, bearing in mind the relative scales of the originals which had to be adjusted.

Thus it is possible to opine that the 1511 and 1512 Western Mediterranean Sea charts are taken from the same pattern/template.


ChAVM/3/D21 ChAVM/3/D22

Having equalised the scale bars the two charts 1511 and 1512 when overlaid are an excellent match with many areas seemingly identical. Thus Vesconte Maggiolo is using the same pattern/template for both.


We see the same accuracy as above when the scale bars are equalised with insignificant differences in the draughtsmanship.


To compare the 1511 chart of this area to the 1512 chart required that several sheets were joined because of the large difference in scale. Firstly the 1512 chart is a complete Europe/Africa chart with a scale bar having 50 miliaria subdivisions. Vesconte Maggiolo clearly informs us on the chart of this fact; “Sapenny cumo da uno ponto alarty sono milia cinquanta” and this fact allowed the various charts drawn to differing scales to be aligned.

Given the small chart in the Parma 1512 Atlas covering this vast area of the world from 60N to 35S and 35W to 45E it is a masterpiece in its own right and compares well with the JCB 1511 pages I have used to form the same chart.

However given that fact it would be stretching credibility to investigate further.


ChAVM/3/D24 ChAVM/3/D25 ChAVM/3/D26

THE 1519 ATLAS- MUNICH LIBRARY- Cod.icon.135


This atlas comprises 7 sheets and is drawn with the first sheet having a half size scale bar to the norm;”da uno quadro alartio sono mill cento” and “da uno punto alarto sono milia vinty”. That is the lines are 100 miliaria apart and each dot distance from the next is 20 miliaria. The whole atlas is as Diagrams ChAVM/3/D27; D28; D29; D30; D31 & D32/33





ChAVM/3/D32 ChAVM/3/D33

BSB-Hss Cod.icon.135; Sheet 1 analysis

ChAVM/3/D34 ChAVM/3/D35

This chart comprising part of the west coast of Iberia and Africa has the newly discovered South American continent and the West Indies there-on. It has appended a vertical latitudinal scale with a scale bar having 100 miliaria subdivisions. From that fact we can establish that each degree of latitude is drawn as 75 Miliaria and thus the standard problem with Portolan Charts is revisited. One degree of latitude should be 90 Miliaria as drawn on most Portolan Charts east of the Pillars of Hercules. But the west coast of Iberia is always drawn having its 6 latitudinal degrees, 37N to 43N as 75 Miliaria mimicking the 75 Roman Miles of a degree. Hence as I have already opined many times the origins of the slewing of Portolan Charts begins with this fact of the mis-scaling of Iberia.

Thus the latitudinal scale is 75 Miliaria per degree, not the correct 90 Miliaria per degree, but that is clearly exhibited at the Equator as the distance from 17W, Cape Verde to 10E, Equatorial Guinea clearly illustrates.

It is then possible to relate the West Indies drawing to the whole chart where we find that the South American coast is correctly positioned 35W/5S as is the longitude of Trinidad at 60W. The latitudes are of course awry as Trinidad is 10N and the junction of Hispaniola/Cuba is 20N. Those two lines are actually 1100 Miliaria apart and thus the latitude scale here is 110 Miliaria per degree and not the Standard 90 Miliaria or even the awry 75 Miliaria.

BSB-Hss Cod.icon.135; sheets 2 and 6 analysed

The sheet 2 is a continuation of Africa southwards from Sheet 1 (see later) and also includes the west coast of the Indian Sub-continent.

But sheet 6 has India and the Golden Chersonesus as well as Sumatra (Taprobane) and Java. This combined diagram has the longitudes remarkably accurately drawn and I have extended the Equatorial line to illustrate the latitudinal errors.

It is obvious that sheets 1, 2 and 6 combine to form a complete Planisphere from c90W to c105E and the two diagram sheets illustrate the fact clearly when joined at the specific marks there-on appended. Diagrams ChAVM/3/ D34; D35 and D36 refer.

I would therefore opine that perhaps the order of the sheets in the atlas as extant may not be as originally intended.

BSB-Hss Cod.icon.135; sheet 3

ChAVM/3/D36 ChAVM/3/D37 ChAVM/3/D38


This sheet covers the west coast of Europe and Africa from c60N to 27S and c30W to 0E. It contains no surprises and plots the Azores correctly as is clearly shown.

BSB-Hss Cod.icon.135; sheet 4 and sheet 5


These two sheets form a standard Portolan Chart from the Pillars of Hercules to the Black Sea eastern shore-line. Again there are no surprises, but if these two sheets are joined to sheet 3 then there is a rather magnificent Portolan Chart comprising the Atlantic seabord and the Mediterranean Sea basin.


Thus I think sheets 3, 4 and 5 should precede sheets 1, 2 and 3 as their derivation would be from afore-going.

BSB-Hss Cod.icon.135; sheet 7

A rather well drawn and accurate depiction of the Mare Caspian with its latitudinal measure perfectly scaled to match the geographical distance as the comparisons show.




The base chart is the Munich 1519 sheet and the Parma 1512 plot has been traced with an equalised scale bar indicating that the difference between the two plots is insignificant. Again it is fair to opine that this could not happen unless the same pattern/template was available in both Naples, 1512 and Genoa, 1519.




When I first traced these two charts with equalised scale bars I was quite astounded at the difference that is clearly shown on the first diagram. However it appeared to me to be a matter of the scales not actually being equal as the features were remarkably similar which is as I would have expected from the above comparison. I thus calculated from the first diagram that the Munich Chart scale was some 5to 6% over sized, probably by mis-copying from the scale ruler used to draw the chart. I therefore reduced the chart to 95% original from the equalised chart and re-traced the two. The comparison is shown on the second sheet and the small differences have been drawn for the Parma/Munich charts.

I have often stated that the scale bars when transferred from the scale ruler used to draw the chart to the chart itself is a possible cause of some inaccuracy as in some instances in my research it was obvious a small part millimetre per unit would remove some inaccuracy.



When the scale bars are equalised the two charts are in fact virtually the same with very insignificant differences visible.



The Parma 1512 Atlas pages and the Munich 1519 atlas pages are drawn from the same pattern/template which it appears was available to Vesconte Maggiolo from c1509 to c 1519 and no doubt beyond that date.

SECTION 1; the mathematical reduction in the scales to produce the 7 atlas pages making up the JCB 1511 Atlas are a masterly progression culminating in the use of the Golden Section or Ratio. That raises the spectre of Vesconte Maggiolo being primarily an artist as already suggested by some researchers. The Golden Section or Ratio is certainly an artistic tool used to position the sections of a painting. It was also used by printers to set the text upon the page and as previously indicated was known from the work of Leonardo of Pisa known as Fibonacci in the 13th century.

All I can opine is that to find the Golden Section or Ratio on a chart is no doubt a deliberate act by the cartographic draughtsman.

The use of the Roman Pes and its twelfth part the uncia is known throughout Italy at this time and even made its way to France from the French occupations.

Thus we may add to Vesconte Maggiolo the attributes of those of a mathematician.

SECTION 2; it is obvious that Vesconte Maggiolo arrived in Naples by 1509 at the latest and was then some 35/39 years old. Therefore he had a career prior to his arrival. That career must have included cartographic draughtsmanship as he could not have drawn the 1511 atlas without the aid of a “full” Portolan Chart to dictate and illustrate the atlas chart pages. He must therefore also have carried with him a pattern/template to enable future work. Whether this is the “1504” Fano Chart which is the subject of much debate regarding its dating is open to discussion, but it must be shown to be a very viable base chart for the atlases which follow.

The fact that the 1511, 1512 and 1519 Atlas pages although varying widely in scale are also so very close in their content, the drawing of the coast lines etc., indicates that the same pattern/template was available from c1509 to 1519 at least.

It is also apparent that the 1519 Atlas probably laid the foundation for the later 1531 Planisphere which will be the subject of another text when the whole oeuvre of Vesconte Maggiolo is discussed.

That Vesconte Maggiolo was an accomplished cartographic draughtsman may be in part attributed to his possible “artistic” beginnings and the fact that he appears also to be an accomplished mathematician should not be ignored.



The cartographer Salvat de Pilestrina is basically missing from the texts because of a lack of historical records prior to his being noted as drawing the 1511 chart, the first extant chart which may or may not have been drawn in Majorca as will be investigated in this section of text. In the various texts which mention Salvat de Pilestrina there have been many attempts to attribute “unknown authorship” charts to him. These include the two charts known as “Kuntsman II and III” held in the Munich library and they are speculated as being from 1502/06 and 1503.

That he was known on Majorca is attested in Notarial records which D Julio Rey Pastor and Ernesto Garcia Camarero carefully examined and described all in their 1959 text entitled, “La Cartografia Mallorquina”. There is a second Chart by Salvat de Pilestrina attributed/signed and dated 1533, held in the Biblioteca of Toledo.

However Leo Bagrow in his “History of Cartography” makes the following comment; “It appears that the reputation of these Catalan Maps led many Italians, among them Dalorto, Petrus Roselli and Salvat de Pilestrina (in the 14th, 15th and 16th C) to go to Majorca to study the art of her cartographers. Many Catalans also settled in Italy and continued their work there.”

Thus there is the greatest possibility that Salvat de Pilestrina was Italian and probably trained by one of the listed cartographers from around 1490/1500.


The 1511 chart by Salvat de Pilestrina is shown on diagram ChAVM/3/D45 taken from the BnF Gallica website and is obviously an up-to-date standard portolan chart with features taken from a variety of Mallorcan Charts, particularly Petrus Rosselli.

The chart is endorsed; “Salvat de Pilestina en Mallorques en lav M.D.xi” which is written in the Mediterranean lingua Franca with Italian tinged patois.


But did it begin its life elsewhere and was carried as a partial chart to Majorca?

We can answer that question as follows as there is one very important feature which is not actually drawn normally on Majorcan Charts and that is the nearer correct portrayal of both British isles, Britannia and Ireland.

Having just completed, as the foregoing texts ChAVM/1/and ChAVM/2 attest, the analysis of the JCB 1511 Vesconte Maggiolo Atlas it was so very apparent that the drawn form of the British Isles was practically identical on the Atlas sheet 4 and the Portolan Chart sheet of Salvat de Pilestrina.

The first diagram, ChAVM/3/D46, is a simple equalising of the scale bars of the 1511 chart and the 1511(1510) Atlas page of the British Isles and setting them side by side to illustrate the fact that they are virtually identical in their draughtsmanship.

The second diagram ChAVM/3/D47 is an overlay of one chart on to the other which illustrates that in many positions it is not possible to differentiate the one from the other.

Perhaps not the wisest methodology, given that the Atlas page is c39 x 57.6cms and the chart is c89 x 75cms, thus it is in area three times the Atlas Page, but it is all that is available to gauge the proof that these are one and the same base pattern/template.

The diagrams are ChAVM/3/D48; D49; D50 and D51, with overlaid charts.

Hence by equalising the scale bars from the Atlas Portolan Chart page to the Salvat de Palestrina Chart it was possible to trace off the four quarters, not trying to align the whole chart as that would be a ludicrous exercise given the disparity of size, but to indicate they were one and the same chart, drawn from the same pattern/template albeit at very different scales.

That of course leads to the simple fact that Salvat de Pilestrina was in Genoa, probably either trained by Vesconte Maggiolo or even trained with Vesconte Maggiolo by Albino de Canepa, as my text ChGEN/1 illustrates.

ChAVM/3/D48 ChAVM/3/D49 ChAVM/3/D50 ChAVM/3/D51

The historical timeline is as follows; Vesconte Maggiolo is born c1470/75 and trained by Albino de Canepa, 1480/85 to 1490/95, and then worked as possibly both an Artist and Cartographer until c1508 when he departed for Naples and drew his JCB Atlas in 1510, finishing it in January 1511.

If the first chart by Salvat de Pilestrina was started or even drawn completely in Majorca, 1511, from the same pattern/template that Vesconte Maggiolo was using it could only have originated in Genoa. Thus the likelihood is that Salvat de Pilestrina was in Genoa 1490/1500 and departed for Majorca, perhaps as Vesconte Maggiolo departed for Naples taking the basic chart with him. Genoa was not then a profitable City for Cartographers, there appears to be only one at any given time as my text ChGEN/1 adequately proves and it would have been a good career move for them both, particularly for Vesconte Maggiolo as the records indicate.


Salvat de Pilestrina is an Italian, probably from the town of the same name south west of Bologna, travelled to Genoa where he leant the art of cartography and then travelled to Majorca to establish himself using their decoration on a basic Genoese Chart former.