Province of Hispanian Baetica


The final form of the province of Baetica was created, by Augustus, in 27BC. from what had been, up until then, part of the province of Hispania Ulterior and assigned to the Senate. It then consisted of more than 175 inhabited settlements and was divided into four judicial sections.

Outline of the Province

In §2/8, Ptolemy sets out the boundaries of the province, tri–angular in shape, indicating that, on the western and northern sides it shares a boundary with Lusitania and Tarraconensis respectively and on the eastern and southern sides it shares the continuation of the boundary with Tarraconensis and the Outer and Balearic Seas respectively. The cardinal points given are;

Ptolemy's coordinates as opposed to those of today

South/West – mouth of Anas river (Guadiana) 4°20¢ 37°30¢ 10°36¢ 37°12¢

North/East – location of this river where it

crosses into Tarraconensis. 9° 0¢ 39° 0¢ 13° 7¢ 39°22¢

South/East – Promontory of Charidemi. 12° 0¢ 37°15¢ 15°28¢ 36°52¢

At first glance there would appear to be no relationship at all between Ptolemy's co–ordinates and those of the present time except to note that there is a severe longtitudinal drift to the west in his readings.

Inhabited settlements

In the remaining paragraphs, §9/16, Ptolemy describes the inhabited settlements tribe by tribe. But, of the one hundred and seventy–five settlements included in the creation of Baetica by Augustus, Ptolemy lists eighty–eight and of these only twenty–one can be reasonably identified to known Roman sites by the similarity of their names. These, with their present day co–ordinates, are,

Long. Lat.Long. Lat.
Corduba Metropolis9°20'38°05'13°14'37°53'
Artigis (Astigi)9°40'37°20'12°56'37°33'
Nabrissa (Nebrissa)5°40'37°30'11°50'36°55'
Asta (Hasta)6°00'37°20'11°50'36°48'
Ispalia Metropolis (Hispalis)7°15'37°45'12°01'37°24'
Baisippo (Baesippo)7°1537°05'12°05'36°12'
Gadeira (Gades)5°40'36°30'11°42'36°32'

Tribal Groups

Ptolemy mentions four tribal groups, the Bastuli which he notes are Phoenician by descent, the Turduli and the Turdetani. There is also a region noted as 'ancient towns of Celtic Baetica' but no indication as to whether this region is inhabited by a Celtic or Celti–Iberian ethnic grouping. As might be expected the Phoenician Bastuli inhabit the coast south of the river Guadalquivir right around to the border with Tarraconensis. North of that river the Turdetania are situated. on the coast and along the border with Lusitania and inland from the south coats and along the border with Tarraconensis are the Turduli.

A summary of the commentary for all the provinces of Hispania is made following that of Tarraconensis.

Province of Hispania Lusitania

The province of Lusitania was created by Augustus, in a similar fashion to Baetica yet its commercial importance was not developed until much later. Again, as with Baetica, it is roughly triangular in shape, the northern side following the course of the Douro river and then turning southwards to a high point in the Sierra Gredos. From there it proceeds to the headwaters of Guadiana river which it follows down to the Atlantic Ocean coast. To the west it is bounded by that same ocean.

In paragraphs §1/4 Ptolemy outlines the province along these lines and the co–ordinates for the cardinal points are,

Douro (Dourius) river mouth 5°20¢ 41°50¢ 9°23¢ 41° 9¢

Guadiana (Anas) river mouth 5°40¢ 37°45 8°36¢ 37°12¢

Sierra Gredos (Tagus source) 10° 0¢ 40°10¢ 12°50¢ 39°48¢

Once again it may be seen that the longtitudinal readings are far to the west of the presnt day, while the latitudinal readings are a shade to the north.

In paragraphs §5/9 Ptolemy lists, by tribal groupings, the inhabited settlements of which there are fifty–eight. Only fourteen of these can be identified with known Roman locations on the evidence of name only. These are,

Long. Lat. Long. Lat.
Pax Julia5°20'39°00'10°08'38°01'
Julia Murtilis (Myrtilus)5°15'38°45'10°20'37°18'
Aiminion (Aeminium)7°20'41°30'9°35'40°12'
Scalabisos Colonia (Scallabis)6°00'41°00'9°20'39°14
Norba Caesarea7°50'39°55'11°37'39°29'
Augusta Emerita8°00'39°30'11°40'38°55'
Caicilia Gemellinan or Metelina8°30'39°30'12°02'38°58'

Like the co–ordinates for the outline, the longtitudinal readings show a westwards displacement and the latitudinal readings a slight increase to the north.

The tribal groupings that Ptolemy assigns to Lusitania are the Tordetani along the Atlantic coast towards the southern end, alongside these in the interior, an unknown Celtic tribe for whom there are no Roman settlements. Above these and forming the largest ethnic group, the Lusitani and over towards the border with Tarraconensis, the Vettones.

Province of Hispania Tarraconensis


The province of Tarraconensis is oddly shaped and can most easily be resolved into a very rough triangular shape, into the base of which a massive bite has been taken. This leaves an area for which there are in fact five cardinal points; paragraphs §1/20 describe these locations.

Ptolemy Today

. Douro river mouth 5°20¢ 41°50 9°23¢ 41° 9¢

. La Coruna 7°15¢ 45° 0¢ 9°36¢ 43°22¢

. Pyrennes (Biscayan end) 15°10¢ 45°50¢ 16° 8¢ 43°20¢

. Pyrennes (Balearic end) 20°15¢ 42°20¢ 21° 9¢ 42°25¢

. Gulf of Almeira 12° 0¢ 37°15¢ 15°32¢ 36°41¢

It will be noted that Ptolemy's readings continue to observe the same drift to the west and north that was demonstrated in Baetica and Lusitania

Ptolemy lists about two hundred and eighty inhabited locations in Tarraconensis, of which only a fraction can be identified by name with known Roman locations. The following list is by no means complete but serve to indicate whether the constant error factor is also present in the inhabited locations.

Long. Lat. Long. Lat.
Flavia Brigantium7°15'45°00'11°36'43°22'
New Carchendon (Carthago Nova)12°15'37°55'17°01'37°36'
Iria Flavia6°25'44°30'9°20'42°44'
Lucus Augusti7°25'44°25'10°27'41°00'
Astorica Augusta9°30'44°00'11°56'42°37'
Uxama Barca13°00'44°15'14°5742°52'
Clunia Colonia11°00'42°00'14°36'41°45'
Uxama Argeli11°30'42°30'14°56'41°35'
Satabicula (Saltigis?)13°40'38°55'16°16'38°56'
Illicias or Illicis12°20'38°30'17°19'38°16'
Caesarea Augusta14°30'41°30'17°06'41°39'
Sergontia Paramica14°30'43°15'15°22'41°04'
Basconton or Casconton15°00'42°45'16°20'42°00'


As in the remainder of Hispania, these locations of Ptolemy display a considerable drift to the west in regard to longtitude and, in the majority of cases, a slight drift to the north is respect of latitude. There is, however, no consistenty in the difference between longtitudinal readings that might indicate a standard error pattern. What does seem to emerge is that the expected cumulative error to the west is further increased by some unknown exponential factor not subsceptible to any rational explanation.

Ptolemy lists over fifty tribal groupings, ranging from large groups, such as the Astures, Cantabri, Callaici to what amounts to single city/state groups along the sea coasts such as the Bibali, Limici, Trouvi together with smaller Celto–Iberian groups in the interior. Over such a large area of land, separated by diverse mountain ranges that form isolated enclaves, such a pattern of population is only to be expected.


In Ptolemy's eyes Hispania, unlike Hibernia and Albion, must have represented a long estabilshed region in which Rome had been pre–eminent for almost three hundred years and which had been a commercial target for the previous millenia. The majority of the locations in this region would have been long established trading destinations and their precise positions well known. Whatever Ptolemy's quarrel with Marinus in the apportionment and resolution of exact navigational co–ordinates from the distance travelled, as noted in Book I, one would expect there to have been a standard error relationship. For instance, while Ptolemy would accept the distances given by Marinus, his calculation of the number of degrees and minutes attributable to those distances might differ. But always by the same factor. However, if Ptolemy chose to introduce another factor, such as an allowance for deviations etc. as he also mentions in Book I, then a random element would have been introduced that makes it difficult for a present day comparison to be made with the navigational co–ordinates given by Ptolemy.

By the time of Tiberius, many of these locations had achieved the status of ius Lattii; by the time of Ptolemy the whole region had been thoroughly Latinized and introducing new Romanized settlements or changing the names of old settlements. If we contemplate a much earlier date for Marinus, see discussion above, then must assume that he would have seen this region as a large settled region which stretched unbroken along the south coast of Spain and he would have identified it as a place long settled by Greek and Phoenician communities, feeding an hinterland of more primitive communities. Therefore, when examining the Hispanic locations given by Ptolemy, we would expect to find a mixture of fairly recent Roman settlements imposed on an ancient landscape of long established Phoenician trading points and small Greek city states. Any locations where Ptolemy updated the references of Marinus, or inserted new data of his own, might therefore be expected to be fairly obvious and apparent.

The majority of the Hispanic locations, in all the three provinces, when compared with present day readings, exhibit a pronounced longtitudinal drift towards the west and a slight latitudinal drift towards the north. We would, of course, expect to find a basic difference since the coefficient of Posidonius resulted in the value distance of one degree being set at minus 16% and that of Eratosthenes at plus 11% of today's values. Either of these would cause a cumulative error resevoir the further one moved from the established mean or basic grid. While Ptolemy numbers his longtitudal co–ordinates from a grid line established in the far west and his latitudinal co–ordinates from a similar grid line in the south, he does not indicate where his navigational mean is intended to be. One would assume that the equator would serve for latitude and somewhere in the region of Alexandria for longtitude. Against this it must be remembered that, in the ancient world, the equator represented the extreme south of the known world and for the purpose of navigation would have been of little practical use. In regards to longtitude, although Alexandria may have been the scientific centre from Hellenistic times, for the much older Phoenician scientific experience, it would have been a very recent establishment on what had hitherto been an unremarkable portion of the Nile delta.

In Book I, Ptolemy comments that navigators tended to use the latitude of Rhodes, 36°0¢, as a reference since it 'formed the centre of the known world', and while he makes no reference to a mean point for longtitude it seems a reasonable assumption that it was also based on the meridian passing through the very ancient trading port of Rhodes. As regards the latitude, if Ptolemy had, in fact, used the equator as his latitudinal mean we would have expected to find a greater cumulative difference between his Hispanic readings and those of the present day. However, since Calpe (Gibraltar) on the southernmost tip of Hispania, is just above the latitude of Rhodes, no great difference in latitudinal readings has begun to develop. In the same context, the cumulative build–up of the longtitudinal westward drift that is evident in the Hispanic readings, equate reasonably with a mean located at Rhodes. However, it should be borne in mind hat Alexandria or Tyre might equally have served as the mean meridian of longtitudinal measurement to the west.

It should be possible, if Ptolemy is at all consistent in his transmission of data from Marinus, to construct an algorithm for each of the two coefficients, Posidonius or Eratosthenes, and by applying either of these to the co–ordinate given by Ptolemy for a particular location, arrive at a value that should approximate to the present day reading. As well as confirming known sites, such information would be available for identifying the unknown locations. Such an algorithm would need to recalculate the present day longtitude, of the location in question, as though the Greenwich Mean were relocated on the longtitude of, say, Rhodes. This reading of longtitudinal degrees and minutes, should then either be multiplied by 1.16% for the geodesic time base of Posidonius, or by .89% for that of Eratosthenes. However, in order to make this new reading comparable with the corresponding reading given by Ptolemy, this new reading must be subtracted from the longtitudinal reading for Rhodes, since Ptolemy registers his readings from the meridian of the Fortunate Islands, rather than the common mean. A spot test on the location of Calpe, using the Posidonius algorithm, does in fact indicate that such a comparison can be obtained within a few minutes of the present day reading for Gibraltar. However. many more of the locations in Hispania need to be tested to establish this as a fact, a task that is beyond the remit of this present work. If the subsequent use of such an algorithm should in fact be proved correct then, as mentioned above, its manipulation to identify unidentified locations in Ptolemy's work would certainly be of interest. It should be further noted that a similar pair of algorithms would need to be established for latitude, using the parallel of Rhodes as the mean meridian. These algorithms would be simpler, in that Ptolemy measures his latitudes directly from his common mean and there is no need for a subtraction from his base.