The purpose of the Mycenaean Atlas is to furnish accurate lat/lon pairs for Bronze Age sites in the central and eastern Mediterranean. The emphasis of the Atlas is on the Later Greek Bronze Age (the Mycenaean) although the Atlas does include sites associated with the Early Bronze, Middle Bronze, Sub-Mycenaean, and the Geometric. Presently there are more than 4300 named and located sites in the Atlas.
I often include quotes or partial quotes to support the find type or the ceramic horizon (period) sections. Occasionally taking just a part of a sentence eliminates the pronomial referrent or it leaves a point obscure. When that happens I add the missing information in parentheses and include my initials. It looks like this:
If you want a full list of these additions you can search for the string RHC in the search box.
From the control screen you can reach all the available controls in the MAP.
It contains the usual menu controls: Search, Info page, Using the MAP, and Blog.
Every page displays an html textsearch box. You can search for any string in the database by using this box. This string might be substrings which appear in site names or identifiers, types, periods, and in the general note section of the place key report. The following searches are examples of what can be done:
A search for the word 'Tiryns' returns this search results page:
By inspecting the results you can see that each returned result is paired with a link to the place key (or feature key) page on which that tag was found. Sections of the place key page are indexed separately and those sections appear separately on the search results page. Currently the sections are 'Site (Place) Name', 'Site Comment', 'Type Comment', 'Period Comment', 'Biblio Page Title, 'Feature Name', 'Feature Region', and 'General Note'.
So the result Tiryns: Lower Town' is paired with the site identifier 'C845' which, when clicked, returns the Place Key Report for 'Tiryns: Lower Town' as seen in the next image.
The search page allows you to search by lat/lon pair. If you enter such a pair the search page will return a map with your lat/lon pair at the center. The map will display all features and sites within the frame size.
The arguments to use here (in order!) are as follows:
You must always supply a lat/lon pair but you can also specify the zoom level, framesize, both, or neither.
nn.nn indicates latitude or longitude values; latitude comes first.
z[z] indicates a 1 or 2 digit zoom integer values from 2 through 18
0.f[f] indicates a frame size from 0.01 to 0.3. This includes such values as 0.08, 0.14, 0.25, etc.
Do not enter html tags or special characters.
The 'Combined Control' allows you to draw a map using one or more of three criteria: Region, Ceramic Horizon, and/or (artifact) type. To reset these three criteria to null simply click on the 'Reset' button.
Using the control is simple. Click on the arrow of the desired criterion (e.g., 'Region') and you will get a drop-down menu. Click on, say, 'Magnesia'. Then click on the 'Generate Map' button. Your map will appear. Here is the Magnesia example.
It may be that you are searching for sites by region only. In that case it will be simpler to use the new map-based region search box which is illustrated here:
You may, for example, want to see all the sites in Siteia which is the region comprising eastern Crete. Click on the leftmost thumbnail map to bring up the outline map of Crete.
When you click on the label 'Siteia' a map of that region will appear just the same as though you had selected it from the combination control itself. It looks like this:
The Elevation control allows you to draw a map with just those sites (of whatever kind) which lie between two elevations. For example you may be interested in a map of sites close to the sea. To generate such a map you can specify 0 (all measurements in meters) for the Low box and, perhaps, 100 for the High box. When you press 'submit' your map will be drawn.
At the time this guide was written such a search would return 1854 sites. Here's a portion of that map taken from the south-western Peloponnese.
There's actually quite a lot of information on this map. Notice how so many of these sites are right on the shore. It would be easy, for example, to generate a list of these communities just by pressing the .kml button on the map (described later). This map also suggests where the seashore might have been in antiquity particularly in the Pamisos region .
You don't have to use 0 as the 'Low' parameter. You may be interested in just those sites between 100 and 110 m. in elevation. In that case enter '100' for the Low parameter and '110' for the High.
This control allows you to search for habitations in user-defined ranges. The units are square meters; one hectare is 10,000 square meters. At the time this was written the average habitation size (arithmetic mean) in the Atlas is 31900 square meters (3.2 ha.) but this measure is skewed by very large sites at the upper end. A better average measurement might be the median of all habitations (that measure right in the middle of the whole set); that is about 12000 square meters (1.2 ha.). The habitation sizes in the DB are not normally distributed; they are skewed to the left.
I have been told by colleagues that these habitation size estimates are not very reliable. I'm sure that that's true. Putting that question aside for a moment I feel that important analysis can be performed even if all we have are relative sizes of sites. There is a great deal of interest in settlement hierarchy sizes (relative sizes) as a clue to the organization of a society and I know of no other way that professionals have easy access to such data and for that reason I include it.
How accurate are these estimates? How is it possible to derive an accurate idea of a habitation size from a scatter of sherds (which is usually how it's done)? No such estimate can be exact in the commonly accepted sense. It's all a matter of probabilities. Let's say that there is a site called 'alpha' which, when it was inhabited, was exactly and precisely 2.0 ha in size. Alpha has been deserted since LHIIIC and, aside from a few sherds, very few traces of it remain. Now let's say that four scholars, A, B, C, and D visit the site at different times and each gives a size estimate. Those estimates are:
A: 1.2 ha. ('obviously')
B: 3.0 ha. ('A significant site whose size scholar A underestimated')
C: 1.6 ha. ('Fashionable in B's time to overestimate hab. sizes')
D: 2.1 ha. ('..an unimportant site to which B gave exaggerated significance')
The average of these estimates is 1.975 ha. Surprisingly close to the real value of 2.0 ha. Now this example was cooked up in order to illustrate the idea. But does this work in real life? Can a series of estimates converge on the actual value? In lots of cases it can. To see why let us imagine a fifth scholar, E, who visits the site and gives an estimate of 100 ha.
Impossible? Well, nearly. The odds of any experienced and reliable scholar, such as E, giving an estimate such as 100 ha. and which wildly varies from the true value (of 2.0 ha.) are almost exactly zero (0.0). And the odds are also nearly zero for any scholar to estimate its size as 20 ha., or 10 ha. The odds are better that some enthusiastic researcher will return an estimate of 5.0 ha. or 4.0 ha. But experience (scholars A through D) has shown that the most likely estimate of the habitation size of A is going to fluctuate around its true value of 2.0 ha. Of course field conditions can actively work against this idea. There are a number of sites on the seashore (or actually underwater such as Nisakouli (C1921) in Messenia) where the ocean has worn away significant parts of the site. Nonetheless the truth is in the aggregate. If the Atlas only supports relative site size comparisons it will have accomplished its job.
The Controls Page supports a control that allows you to see Atlas coverage of several well-known sources and gazetteers by contemporary scholars. Making a selection and pressing 'Select' will generate a map which will show just those sites (that are in the database) from that source or gazetteer.
So, for example, if you selected 'Alin 1962' and pressed 'Select' then you would be shown a map consisting of DB points which use Per Ålin's Das Ende der Mykenischen Fundstätten auf dem Griechischen Festland, from 1962 as a reference.
Here is part of such a map. Altogether some 407 sites in the database use Ålin's work as a reference. This illustration is zoomed in to just show those in the Argolid
Such a feature allows you to check the coverage in the database for specific important gazetteers. Some are only partly covered. Others are covered as a whole.
Currently the sources supported are:
Ålin, Per, Das Ende der Mykenischen Fundstätten auf dem Griechischen Festland. Carl Bloms Boktryckeri A.-B., Lund, 1962.
Åström. Paul, 'Mycenaean Pottery from the Region of Aigion, with a List of Prehistoric Sites in Achaia', Opuscula Atheniensia, V, pp. 89-110, C.W.K. Gleerup, Lund, 1965.
Banou, Emilia. Beitrag zum Studium Lakoniens in der mykenischen Zeit, Albert-Ludwigs-Universität zu Freiburg. 1996.
Bulletin de Correspondance Hellenique Online. 'http://chronique.efa.gr/'
Boyd, Michael J., Middle Helladic And Early Mycenaean Mortuary Practices In The Southern And Western Peloponnese. np. 2001.
Fachard, Sylvian, La Défense du Territoire; Étude de la Chora Érétrienne et de ses Fortifications. InFolio Editions, CH-Gollion. 2012. ISBN: 978-2-88474-410-2.
Farinetti, Emeri. Boeotian Landscapes; A GIS-based study for the reconstruction and interpretation of the archaeological datasets of ancient Boeotia. Faculty of Archaeology, Leiden University. 2009.
Foley, Anne. The Argolid 800-600 B.C.: An Archaeological Survey : Together with an Index of Sites from the Neolithic to the Roman Period Göteborg : P. Åströms Förlag, 1988.
Fossey, John M., Topography and Population of Ancient Boiotia, Ares publishers Inc., Chicago, Illinois. 1988.
Giannopoulos, T., Die Letzte Elite der mykenischen Welt: Achaia in mykenischer Zeit und das Phänomen der Kriegerbestattungen im 12.-11. Jahrhundert v. Chr., Bonn: Habelt. 2008.
Hood, Sinclair and David Smyth, 'Archaeological Survey of the Knossos Area', The British School at Athens. Supplementary Volumes, No. 14, Archaeological Survey of the Knossos Area, pp. i-x & 1-69, 1981
Hood, Sinclair and Peter Warren, Gerald Cadogan, 'Travels in Crete, 1962', The Annual of the British School at Athens (59), pp. 50-99, 1964.
Howell, R. "A Survey of Eastern Arcadia in Prehistory", The Annual of the British School at Athens. 65. November, 1970. 79-127.'
Immerwahr, Sara A., The Athenian Agora; Vol XIII; The Neolithic and Bronze Ages. The American School of Classical Studies at Athens, Princeton, New Jersey. 1971.
Jameson, Michael H, Curtis N. Runnels, Tjeerd H. van Andel, A Greek Countryside; The Southern Argolid from Prehistory to the Present Day, Stanford University Press, Stanford, California. 1994.
Kanta, Athanasia, The Late Minoan III Period of Crete. Paul Åströms Förlag, Göteborg. Sweden. 1980.
Waterhouse, Helen and R. Hope Simpson, "Prehistoric Laconia: Part I", The Annual of the British School at Athens, Vol. 55, pp.67-107, 1960
Waterhouse, Helen and R. Hope Simpson, "Prehistoric Laconia: Part II", The Annual of the British School at Athens, Vol. 56, pp. 114-175, 1961.
Langohr, Charlotte, PERIPHEREIA: Étude régionale de la Crète aux Minoen Récent II-IIIB (1450-1200 av. J.-C.) 1 | La Crète centrale et occidentale, Presses Universitaires de Louvain, Belgium, 2009.
Lolos, Yannis A., Land of Sikyon: Archaeology and History of a Greek City-State, Hesperia supplements, 39. Princeton: American School of Classical Studies at Athens, 2011.
McDonald, William A. and George R. Rapp, Jr., The Minnesota Messenia Expedition: Reconstructing a Bronze Age Regional Environment, University of Minnesota Press, Minneapolis, Minnesota. USA. 1972.
Melas, Emmanuel, The Islands of Karpathos, Saros and Kasos in the Neolithic and Bronze Ages, Paul Åströms Förlag, Göteborg 1985.
McDonald, William A. and Richard Hope Simpson. 1961. 'Prehistoric Habitation in Southwestern Peloponnese'. American Journal of Archaeology. Vol. 65, No. 3 (Jul., 1961), pp. 221-260.
McDonald, William A. and Richard Hope Simpson. 1964. 'Further Exploration in Southwestern Peloponnese: 1962-1963'. American Journal of Archaeology. Vol. 68, No. 3 (Jul., 1964), pp. 229-245.
McDonald, William A. and Richard Hope Simpson. 1969. 'Further Explorations in Southwestern Peloponnese: 1964-1968'. American Journal of Archaeology. Vol. 73, No. 2 (Apr., 1969), pp. 123-177.
Mountjoy, Penelope A., Mycenaean Athens. Paul Åströms Förlag, Jonsered, Sweden. 1995.
Myers, J. Wilson and Eleanor Emlen Myers, Gerald Cadogan, The Aerial Atlas of Ancient Crete. University of California Press, Berkeley, Los Angeles. United States of America. 1992.
Nowicki, Krzysztof. Defensible Sites in Crete; c. 1200 - 800 B.C. (LM IIIB/IIIC through Early Geometric), Aegaeum (Annales d'archéologie égéenne de l'Université de Liège et UT-PASP), 21. 2000.
Panagiotakis, Nikos, 'L'évolution archéologique de la Pédiada (Crète centrale) : premier bilan d'une prospection', Bulletin de Correspondance Hellénique (127:2), 327-430. 2003.
Papadopoulos, Thanasis J., Mycenaean Achaea; Part 1: Text. Paul Åströms Förlag, Göteborg, Sweden. 1979. Vol 1. 1979.
Pelon, Olivier. Tholoi, tumuli et cercles funéraires; Recherches sur les monuments funéraires de plan circulaire dans l'Égée de l'Âge du Bronze (IIIe at IIe millénaires av. J.-C). Bibliothèques de l'École française d'Athènes et de Rome - Série Athènes, 229. 1976.
Phialon, Laetitia. L'Émergence de la Civilisation Mycénienne en Grèce Centrale, Peeters, Leuven - Liege. Belgium. 2011. ISBN: 978-90-429-2585-4
These entries were taken from PRAP's website. That site is maintained by Sebastian Heath and is here: http://classics.uc.edu/prap/static/sites_list.xsl.html
Privitera, Santo. Principi, Pelasgi, e pescatori: L'Attica nella Tarda Eta del Bronzo, Paestum: Pandemos. 2013.
Sackett, L.H., and V. Hankey, R.J. Howell, T.W. Jacobsen, M.R. Popham, 'Prehistoric Euboea: Contributions toward a Survey', The Annual of the British School at Athens, 61, pp. 33-112, 
Schallin, Ann-Louise, Islands under Influence: The Cyclades in the Late Bronze Age and the Nature of Mycenaean Presence. Studies in Mediterranean Archaeology (SIMA), no. 111. Jonsered 1993. ISBN: 978-91-7081-098-5
Simpson, Richard Hope. Mycenaean Greece. Park Ridge, New Jersey: Noyes Press, 1981.
Simpson, Richard Hope. Mycenaean Messenia and the Kingdom of Pylos. Philadelphia:Instap Academic Press, 2014.
Simpson, Richard Hope and O.T.P.K. Dickinson, A Gazetteer of Aegean Civilization in the Bronze Age, Vol. I: The Mainland and the Islands, Paul Åströms Förlag, Goteborg. 1979.
Souyoudzoglou-Haywood, Christina, The Ionian Islands in the Bronze Age and Early Iron Age; 3000-800 B.C., Liverpool University Press. 1999. 978-0853236542.
Syriopoulos, Konstantinos, He proistorike katoikesis tes Hellados kai he gennesis tou Ellenikou ethnous, B' (The prehistoric settlement of Greece and the origin of the Greek people), Athens: Athens Archaeological Society. 1995.
van Wijngaarden, G., Use and Appreciation of Mycenaean Pottery in the Levant, Cyprus and Italy (ca. 1600 - 1200 BC). Netherlands Organisation for Scientific Research and Amsterdam University Press. 2002.
Zavadil, Michaela. Monumenta: Studien zu mittel- und späthelladischen Gräbern in Messenien. Wien:Osterreichische Akademie der Wissenschaften Philosophisch-Historische Klasse Denkschriften. 2012.
Currently the Helladic material is covered completely for
The others are not completely covered. Much material is redundant. For example Simpson & Dickinson  has only recently been completely integrated because much of its material is duplicated in Simpson  which was completely integrated long ago.
The total number of bibliographic sources for this database now numbers some 1650 items but those gazetteers and monographs listed here are the most heavily used.
In addition to displaying Mycenaean sites on maps this site also provides reporting facilities.
The report page is triggered both from the Controls Page Combo Map Box as well as the Combination Map itself. When, from the Controls page, you select some criterion such as the region name ‘Salamis’ you can then click on the ‘Report’ button.
The resulting REPORT page provides a site breakdown by the number of times sites fitting your criteria are tagged with specific types. This gives you an idea of the dominant use of your chosen sites. There is also a site breakdown by regions and subregions. In this example using ‘Salamis’ all the sites are in the Saronic region and the Salamis subregion.
If you selected some criterion other than region you will get a table that lists all the regions containing sites that fit that criterion. For example if you chose the type ‘Fort’ the table would show all the regions that contain a site typed as ‘Fort’. Here those include ‘Boeotia’, ‘Thessaly’, ‘Corinthia’, ‘Locris Opuntia’, ‘Argolid’, and ‘Achaea’.
The basic REPORT page also has a drop-down menu that allows you to select more detailed reports.
Each detail report will now be described.
The elevation report provides several things. The first is a basic statistical parameters section.
- Arithmetic mean of elevations of the sites that fit this criterion. This is the sum of the elevations divided by the number of sites.
- Median elevation. The median is another measure of central tendency. It is that actual elevation which falls halfway between the lowest and the highest elevations.
- Standard deviation
- Skew The skew shows how much the bulk of the observations are displaced from the arithmetic mean. It is an indication of symmetry in the curve – or the lack of it. If the skew is positive then the bulk of the observations are displaced to the left of the mean; if negative then the bulk of the observations are displaced to the right of the mean. If the skew is 0.0 or close to it then the distribution is more or less symmetrical. Skew might be either positive or negative but in interpreting the meaning of any partiular skew I use the absolute value. So the skew parameter is interpreted as follows:
0.0: The distribution is symmetrical
0.0 to 0.5: The distribution is approximately symmetrical
0.5 to 1.0 The distribution is moderately skewed
Greater than 1.0: The distribution is highly skewed
Kurtosis This is a parameter that indicates how much of the population is in the tails as opposed to the shoulders of the distribution. For more information about skew and kurtosis and how to interpret them see this.
The elevation report also provides two graphs. The first shows the frequency of various elevation ranges; that is, it maps the elevations into the frequency domain. The horizontal axis is elevations from 0 to highest elevation in the set. The vertical axis shows the frequency of sites whose elevations are in that range.
The second graph is a straight-forward plot of sites against their elevations in the order of increasing elevation. The horizontal axis is the site names in order of elevation from lowest to highest. The vertical axis is elevations.
These charts are for the LH tholos tombs in Messenia. In the DB there are 81 tombs that fit this criteria. The left-hand bar chart shows that they mostly bunch up at 400 m or lower with a couple of stragglers between 700 and 800 m. The chart on the right (red line) plots the elevations in order. The green line is an exponential curve fitted to the data. The function form is
E = f(x) = e^(ax + b)
The actual coefficients for this specific curve are given at the bottom. Often for elevations this is a good fit although in this case it looks as though a linear fit might work much better if we had thrown out the two highest values. Sometimes the extreme values are not characteristic of the data set and it might be legitimate to get rid of them.
When I perform that experiment in this case (linear fit throwing out the two largest values) I get a linear fit:
E = f(x) = 4.6472x - 1.1266
which has a linear correlation coefficient of 0.9781.
I carried out this calculation using the tools furnished by planetcalc.com. The basic fitting tools are here The results from doing this calculation on PlanetCalc look like this:
Beneath the charts are two tables. The left-most table is a list of all the sites fitting these criteria which have elevations lower than the arithmetic mean. The right-most table is a list of all the sites fitting these criteria which have elevations greater than the arithmetic mean. The columns in both tables are sortable. The place key numbers in those tables, when clicked, will bring up a place key report for that site.
The Chrono report generates a table of occupation times based on your choices from the selection control on the Controls page. When you click on the 'Chrono' button a chronological table will be created for your selection sites. This chrono table lists all the individual sites in your selection criteria on the left-most column. All the ceramic horizons for which your site is attested will be plotted with 'x's in the appropriate ceramic horizon columns to the right. Let's look at an example.
From the controls page I select 'Peak Sanctuary' under the type label. Then I press 'Report'. The main report page is then displayed:
You'll see the new 'Chrono Report' button. When you press that you'll be taken to the Chrono page:
In the left-hand column are the place keys for all the places in your selection set. They are clickable. When you click on one you'll be taken to the specific Place Key Report page for that site.
The second column is the place name. The names in this column are mouseable. When you run the mouse over them the region name will pop up in a tool-tip.
Succeeding columns, 'EH', 'EH I', 'EH II' etc., are the ceramic horizon columns. If there is evidence for your site having existed, been used, etc. during any specific ceramic horizon then there will be an 'x' in the box for that ceramic horizon. Here we can see that C1164, Troullos on Kea, existed in the LH I and the LH II.
The column headers with the ceramic horizon names are mouseable. When you do mouse over them the approximate year range will pop up in a tool tip.
It must be emphasized that these ranges are very approximate and scholars differ over what the proper year ranges should be for any specific ceramic horizon. Please keep this in mind. As I learn better ranges for specific ceramic horizons then I will probably change these tool tips.
The 'x' characters in the table boxes are also mouseable. When you do pass the mouse over them then the ceramic horizon label will pop up. This is intended to make it easier to work with the table when the column headers are out of sight.
When creating this table I faced the choice of making the table pageable (column headers never out of sight) or sortable. I chose sortable as probably the more useful alternative. When you click on the column heads the table will be re-sorted to put your criterion at the top. Because spaces in the cells sort before the character 'x' by default you'll probably have to click twice on any specific column head to get the effect that you want.
There are buttons on the chrono pages that allow you to go to the elevation report page or the gazetteer report page. Those pages now also support buttons that allow you to go to the Chrono page. This should make multivariate analysis more convenient.
The tables, like any .html table, can be selected with the mouse, in whole or in part, and dropped directly onto an Excel spreadsheet. Here is an example of a table from the chrono page dropped onto Excel:
The Gazetteer report is a simple list of all the sites which meet the specified criteria. The criteria themselves are listed on the top of the page. A sortable and clickable list of all the relevant sites follows this. The list is sortable by clicking on the column heads.
The bibliography for the Mycenaean Atlas Project now numbers more than 1300 items. For a long time I have wanted to be able to present the bibliography online in a focused and analytical fashion.
From the beginning it was possible to create a map that showed every location documented by specific scholars. It was possible to create, for example, a map of the Mycenaean world which showed just those sites for which Simpson’s Mycenaean Greece was one of the sources. At the present time one can draw such a map for any one of 40 different scholars.
This is useful, for example, in tracking the actual coverage that the MAP gives to any one of these prominent writers and these maps change as more and more of their sites are given place in the Atlas. Here, for example, is a map based on Emilia Banou's Beitrag zum Studium Lakoniens in der mykenischen Zeit from 1996. These are all the sites for which her work is at least one of the supporting citations.
There is, however, more in the Bibliography than just those prominent scholars. At a minimum it should be possible to produce customized bibliographies. If a user chooses just to look at sites from Arcadia then it should be possible to produce a bibliography of just those writers who are cited in locations from Arcadia. In other words, to produce a customized bibliography of Arcadia in the Mycenaean period. And that goes for any other combination of choices. A customized bibliography for ‘tholos’ and ‘LH’; or a customized bibliography for ‘Forts’, etc.
This is now possible with new dynamic bibliography report. In addition to the other reports that are generated from the report page such as ‘Chrono’ or ‘Elevation’ it is now possible to generate a bibliographic report limited to just those categories the user actually chose from the controls page.
If the user chose ‘Boeotia’ and ‘EHII’ from the control page then pressing ‘Biblio Report’ on the main report page will produce a new page that contains a bibliography of just those authors and sources in the DB who are cited in support of every EHII Boeotian site.
A dedicated bibliography for 'Boeotia' in the 'EH II' looks like this:
The primary feature of this new page is the bibliography table. This table has four columns, 'Author', 'Citation', 'Comment', and 'ISBN'.
Author: This column displays the author name, if known. If there is a link associated with the author's name then clicking that link will bring up a map with just the sites in 'Boeotia' in the 'EHII' that use that author as a supporting citation.
Citation: This column displays the title and publisher (or journal information) of the work being cited. If there is a link in this column then clicking that link will bring the user to the actual text of the work (.pdf or other doc) being cited.
Comment: The comment field displays extra information about the work being cited.
ISBN: This displays the ISBN number (or ASIN or DOI) of the work being cited. This is included, when known, to prevent ambiguity. The ISBN is still the best way to identify the exact edition of the book being cited.
Here's a part of the biblio page for 'Attica':
Here there are ISBN numbers for Simpson and Hagel's Mycenaean Fortifications ... and Privitera's Principi, Pelasgi, e Pescatori because those books are not available online. On the other hand Jeremy Rutter's "Evidence for a Mycenaean Tomb of the Late Helladic IIA period in the Athenian Agora" is online and the page provides a working link to it. Clicking on that link brings up Dr. Rutter's article.
And what about that first column which is headed 'Author'? As mentioned above that entry might have a link and, if it does, clicking on that link brings up a new map with sites for which that author/work combination provide bibliographic support. For example, when you click on Jeremy Rutter's name next to his article 'Evidence for a Mycenaean Tomb of the Late Helladic IIA Period in the Athenian Agora' from Hesperia you are taken to a new map page with just that site on it for which Dr. Rutter's article is cited as support. It looks like this:
Here's the new single citation source page. Since you clicked on Dr. Rutter's name it brings up the single site in Attica which is in the DB and for which Dr. Rutter is one of the sources. In this particular case it turns out to be a grave in the Athenian agora (C1835). In this way you can quickly track those sites for which specific authors are cited as supports.
At the top of this page the original choice criteria are repeated. The Criterion Bar says "Source: Rutter  Number of sites: 1 Region: Attica"
When you mouse over the Source the full citation will pop up. It looks like this:
Here you can see the black background popup box with the full name of Dr. Rutter's article and its place of publication, namely in Hesperia.
Ceramic Horizon (Period) criteria are handled in a similar fashion. Here, for example, is a map derived from Simpson's 1959 article on 'Nemesis'. One selection criterion was 'LH'. For our search purposes the choice of 'LH' means every subdivision of that period: LHI, LHII, LHIII, etc. You can see that if you mouse over the 'Ceramic Horizon: LH' string:
The Reporting facility also provides the ability to examine the aspect of a large number of sites at the same time. You reach the Aspect Map from the main report page like this:
After you have selected the Aspect Report that page will appear. I use here the example of the BA sites in Apulia which is a subregion of Italy.
The Aspect Report allows you to see all the aspects of all the sites in your chosen set. Here I have created an Aspect Report from the sites in Apulia. You will see that the page consists of three main parts:
On the map the sites are represented by directed arrows which indicate the general aspect of the site. If you mouse over an arrow then a popup will appear that gives the site place key and name. If you click on the arrow then an info box will appear that gives more information about the site along with a clickable link that will take you to the Place Key Report page for that particular site. Here is what that looks like in our example of Apulia:
Here I have clicked on C6369 (Masseria Vecchia) which happens to be a place where a Mycenaean pot was found
Before discussing the Place Key Report page it is important to be able to read and interpret the information on the map. The sites on the map are indicated with icons. These icons have the following meanings:
To retrieve a report about a site you search for it by entering the place key ID in whatever search box is handy. The place key is unique to each site. It consists of the letter 'C' followed by a three or four-place number. Here's an example. You can try entering something like 'C219' (without the quotes).
That place key would have taken you to a search results page with a link on it for the place key 'C219' which is the report page for 'Ano Engliano' (Nestor's Palace).
On that report page there is a map of Ano Engliano and its outlying regions. C219 (the palace) is centered on the map. Also there are several sections describing the site.
There is a general information section which gives you the exact latitude and longitude in decimal and DMS form. You are also given the location in Universal Transverse Mercator, and in What3Words form.
You can see from this image clip that you also get the region (Messenia), the elevation of the site (186 m.) and an indicator of how accurate the given location actually is (Accuracy). There are occasions where that is not possible and when that happens the database supplies an error radius. This is radius of a circle, centered at the lat/lon pair, which touches some part of the sought-for feature. For a complete description of this parameter and of the concepts of Precision and Accuracy as they apply to this web site see this. In this case we are certain that the supplied lat/lon pair actually is the location of Nestor's Palace.
The word 'POSITION' on this report is a link to Google Maps. When you click on it Google Maps will open centered on the lat/lon pair in question: Here Nestor's Palace. This is convenient because Google Maps provides a Street View facility which is often helpful when researching sites.
The Type section provides some information about the site type. It may be a tholos, a habitation, or something else. Here is an example from site C445 which is the tholos tomb at Koutsochira: Diasela in Triphylia.
In this example there are three citations for the type. One is from McDonald and Simpson's famous 1961 article about the Peloponnese (described at the bottom of the place key report). McDonald and Simpson typed it as a tholos tomb. In the second citation McDonald and Rapp, in their report on the Minnesota Messenia Expedition emphasized the general nature of the site as a cemetery. In the last citation Simpson, in his 1981 Mycenaean Greece, typed it as a tholos.
The next section provides chronological information about the site. For nearly all sites this is the most difficult information to provide. Ceramic horizons can often be identified (or plausibly guessed at); it is the conversion of these ceramic horizon names into actual year ranges which is controversial. Ordinarily a number of chronological citations are supplied here (with many more to come in the future). It should serve to at least orient you in this area.
You should be aware that the year ranges given are expressed in the 'High' Chronology. The date of the transition from LMIA to LMIB (LHI to LHIIA) is controversial. Both sides agree that the transition occurred shortly after the eruption of the volcano on the Cycladic island of Thera and the traditional date attached to the eruption (based largely on pottery sequences) is ca. 1500 BC. But more and more scientific evidence is accumulating that indicates the eruption occurred about 100 years earlier; perhaps about 1628 B.C. That means that the (LMIA to LMIB) LHI to LHII transition is to be dated about 100 years earlier than used to be thought. The 'scientific' arguments for retaining the traditional transition date are looking more and more strained and so this site adopts the earlier (the 'High') chronology. The most reasoned and reasonable discussion of the issues can be found in Sturt Manning's essay entitled "Eruption of Thera/Santorini" which comprises Chapter 34 (pp. 457-474) of the Oxford Handbook of the Bronze Age Aegean (Eric Cline, ed), OUP, 2010. The Mycenaean Archaeological Project database itself does supply Low Chronology equivalents for the Ceramic Horizon names but this feature has not yet been incorporated into the website.
The following is an example of the Period (Ceramic Horizon) information for site 'C125' which is a site at Mandra in Messenia.
In this example five ceramic horizon citations are supplied from two different works, McDonald and Rapp  as well as a famous article by McDonald and Simpson from 1969 (sometimes called 'Messenia III'). The ceramic horizon name is presented followed by the short form citation of the authority being used. Oridnarily each citation is accompanied by a comment that justifies my citing them in this way. So in citation 1 I have quoted McDonald and Simpson: "A sherd from a thin-walled open shape could be MH or EH", to support the ceramic horizon name of 'EH'. The long form of the bibliographic citation is given in the next section.
In the bibliography section attached to C125 at Mandra four citations are supplied. Each citation gives the short form (used in the rest of the report), here, for the first citation, 'Simpson ',
then the full title: 'Simpson, Richard Hope. Mycenaean Greece. Park Ridge, New Jersey: Noyes Press, 1981.'
followed by the chapter or paragraph name: 'F 235 Mandhra: Chazna'
and, lastly, the page which is 141.
The bibliography section is not meant to be exhaustive. Its purpose is to give you at least one reliable source from which to start your investigation into the site. A fine general bibliographical work in this area is Bryan Feuer's Mycenaean Civilization, Revised Edition, McFarland & Company, Inc., North Carolina, USA. 2004.
The website helladic.info also supports nearest neighbor analysis for every site in the database. This is part of an effort to integrate all sites to their surrounding environments.
A dedicated neighbor page can be accessed from any Place Key Report page. There is a new button, ‘Neighbor’, on the lower menu line of that page. When that button is pressed a new neighbor page will appear based on your anchor site. This page contains new graphs and a table based on analyzing the anchor site’s neighbors.
To perform this analysis the software searches the database for sites within 1500 m., in any direction, of the anchor site. The area covered is 3 km. square or 9 square km. These neighbors are then displayed in the form of a sortable table that lists the place key, name, type, elevation, and distance of the neighbor from the anchor site.
2. Neighbor Chart. Here for neighbors of C5443, Voni: Sochora, which is a town in the Pedhiadha of Crete.
The neighbors are also displayed on a new map at the upper right. This map shows only the neighbors and does not display features.3. Neighbors map for C5443, Voni Sochora.
In addition the neighbors are displayed on two new charts. The first is a bull’s eye chart which depicts the anchor site in the center and the neighbors in the form of concentric circles around it proportionate to their distance from the anchor site. The bull’s eye chart highlights gaps or unusual distances between the anchor site and its neighbors. Also it easily shows neighbor sites that are similar distances from the anchor site (no matter in which direction) because these similar-distance sites cluster on the bull’s eye chart.4. Bull's Eye chart for neighbors of Voni Sochors, C5443 (at center).
The second chart is a vector graph that depicts each neighbor as a vector with distance and length proportional to its distance from the anchor site. This chart is an abstract representation of the neighbor map on this page. The vector chart, however, depicts distance and direction in a highly abstract way that can be taken in at a glance. Each vector is labelled with the place key of the neighbor in question. It will quickly show, for example, whether the anchor site is surrounded by neighbors or whether it is on one of the edges of a cluster of neighbors.5. Vector chart for C5443 Voni Sochora (at center)
Scales have been drawn to make analysis easier. These scales are useful in measuring distances from the anchor site to its neighbors. The scales display tic marks at 100 m (1/10 km.) intervals. The center value (aligned to the anchor site) is marked ‘0’. Positive distances are marked at 1 km with the integer ‘1’ at ten tic marks to the left and also to the right.
For both of these charts and for the map north is to the top.
The user should bear some cautions in mind. Neighbors are, of course, limited to whatever is in the database and this site cannot guarantee that the list of neighbors is exhaustive. Also each site has an associated accuracy parameter. If the accuracy parameter for neighbors is not exact then the real distances could be significantly different.
The user should also be aware that at present the neighbors shown on the map and the graphs are from all time periods and may not have existed at the same time as the anchor site. I'm working on making this more sophisticated.
Also there are some areas for which this approach does not work very well. Here is an example from the Athenian Agora (C1868) where there are so many sites that the graphs basically become unreadable:
Enlarging these charts would not help. The whole thing is just a smear. There are also sites which have no neighbors at all or, at least, none as defined here. Here, for example, is the neighbor page for Larisa on the Hermos (Aeolis) C1473:
For many areas, however, this page should be a valuable aid for the analysis of Bronze Age sites. Here's one more example, from one of the well known (and now disappearing) mounds (C319) at Lefki Kaldamou in Messenia:
The graphs quickly and precisely show that these mounds are more or less evenly spaced and arranged in a nearly straight line.
These charts are implemented using the HTML 5.0 canvas. The canvas construct does not support links. You can, however, link directly to any of these neighbor sites from the table on this page.
Three-dimensional terrain models (landscapes) are available for Ninety percent of the sites. These models are scalable and rotatable and they allow you to get a better feel for the terrain. You reach a site's 3-D landscape from that site's place key report page Toolbox menu. Here, for example, is the place key page for the Cretan 'refuge' site of Karphi.
When the toolbox menu is pulled down you see that there is a clickable choice that says '3D'.
This takes you to a page with your 3D terrain model.
Here you are looking at the terrain model for the area around Karphi. It originally draws in 'live' mode which means that you can manipulate the image with your mouse. The controls are:
When the image is first displayed south is to the bottom (or towards the viewer) and north to the top and away from the viewer. So here, for example, I have used the mouse to re-orient the image so as to be looking at Karphi more from above and from the south-west.
In the next image you can see that the 3D-terrain model is tagged with numbered circles. These are links to the locations (in the database) that are in the Karphi area. Number 1 is just the Karphi Sanctuary (C6241) itself.
If you click on that link a popup window will appear.
The link that you see on the popup window is a link back to the orginal Karphi page (C6241) and clicking on it brings you back to that page.
The concept of the slope of a site is defined as the difference in elevations between two points divided by the distance between the two points (using same units of measure).
The slope is, of course, nothing more than the tangent measure between the two points.
Eight sample elevations are stored for each site in the atlas . These elevations were derived for points N, S, E, and W from the central site at the distances of 150 and 300 m.
For each elevation sampling the slope is the difference in elevations between the central site and the sample. This difference is then divided by either 150 or 300 as appropriate. This is the slope for that elevation sample.
In this atlas site aspect is defined as the direction in which the terrain most rapidly falls away from the site. In other words the site aspect is the direction of the low ground at and near the site. If the ground falls away from the site to the east then the site is defined as having east aspect. For that to be true the slope to the east of the site would be, mathematically, the smallest (or most negative) of the slopes around the site. These mathematical measures often contain a great deal of information. If the aspect of a site is north and west then the site has NW aspect. If it has east and west aspect (slopes smallest or most negative to the east and west) then it would be sitting on a ridge that runs north and south. If a site has negative slope on all four sides (as sites often do) then it would be sitting on top of a hill. Its aspect would be the side(s) with the most negative slope. This is often the case with peak sanctuaries in Crete or Kythera.
For example here is part of the Place Key Report page for the peak sanctuary at Petsofas in Siteia on Crete. Just looking at the map makes us expect that the ground will fall away from this peak in every direction but, perhaps, most strongly to the north and west.
Can we show that mathematically? Sure. The interface provides an Aspect page for each site. To reach that page just pull down the toolbox menu on the Place Key Report page and click on 'Aspect'.
This will bring up the aspect page for Petsofas (C6228).
As you can see there is a bar graph on the left which displays the slopes for each of the cardinal directions at 150 m. (blue green) and 300 m. (grey). In this case all the slopes are negative (0 at the top) which means that the ground falls away in each of the four cardinal directions.
On the right are some tables. The first table gives information about the site, the place key, the place name, and its elevation.
The left side of the chart displays a scale in meters. Shapes that stretch out strongly either up or down alert you to significant slope in that direction.
Always consult the chart scale. Scale intervals may differ between sites and the scale will not always center at 0 m.
The second table relates the site to eight other positions. These positions are just those places which are 150 m and 300 m to the north, east, south, and west of the central position (here the peak sanctuary of Petsofas). The chart displays three columns for each of the two sampled distances, 150 and 300 m, and a row for each of the four cardinal directions. In the first column the position elevation is displayed. The next column displays the slope between the site and that sample.
The last column of each group of three attaches a qualitative description of the slope at that point. Here I follow a protocol found in Malaperdas and Zacharias (1). I quote from their adaptation of this protocol as follows:
|0.0 - 3.5%||Little or None|
|3.5 - 8.7%||Slightly Inclined Relief (Gentle)|
|8.7 - 26.8%||Strong Inclined Relief (valleys, tectonic surfaces)|
|26.8 - 70.8%||Steep to Extremely Sharp Relief (mid-valley slopes)|
|70.0 - 135.0%||Steep Relief (slopes of high valleys, canyon slopes)|
|135.0% +||Vertical (vertical slopes in mountain areas)|
How can these sample elevations be used to help analyze the landscape around a site? Jeroen De Reu and associates illustrate this in an article from 2011. He and his team were analyzing Neolithic barrows in Belgium; to facilitate this they developed three measurements.
The first is elevation percentile (PERC) which is simply the percentage of sample elevations lower than the site itself.
The second measurement (DIFF) is defined as the difference in meters between the site elevation and the median of the sampled elevations.
In order to compute the Z-score or "DEV" this page also presents the SD or standard deviation of the sample elevations.
The third measurement (DEV) is defined as the deviation of the site elevation from the mean elevation. Statistically DEV is the Z-score of the site relative to the distribution of the sample elevations.
In the Mycenaean Atlas the Slope and Aspect page presents all three of these measurements for each site. These measurements must, however, be used cautiously because the meaningfulness of the Standard Deviation relies on the elevations taken at the sample points be normally distributed. But this is an unsafe assumption. Sampled elevation measurements are not ordinarily normal but log distributed. I can think of but one example that I have seen in which the charted elevations appear normally distributed and that is the case of site elevations in Arcadia. In Arcadia most settlements have been on the fertile and well-watered 600-800 m. high plain. There are relatively fewer settlements or sites at elevations below that range or above it. This causes measured elevations in Arcadia to cluster in the middle of a distribution graph and give the graph a normal appearance – one in which Standard Deviations and Z-scores would have some meaning. The next illustration is the elevation report page for sites in Arcadia:
So, in addition to the measurements described above, the Slope and Aspect page analyzes the relative values of the slopes at 150 and 300 m. The smallest or most negative slopes lie in the direction that the site ‘faces’. On the SLOPE and ASPECT page those are emphasized in a separate notice.
In addition, for each cardinal direction, the Slope page presents a simple linear function in the form
E = f(x) = Ax + B
where E is a predicted elevation and A and B are the coefficients of the least squares linear function that comes closest to all three points (site elev, elev at 150m and elev at 300m) in EACH of the four cardinal directions. The B coefficient in this function is an approximation to the elevation of the origin site. Evaluating the function, f(x), for 150 and 300 gives approximations to the elevations at the sampled distances.
I was motivated to perform this analysis because I believe that the coefficient A, the slope of the fitted line, might also be a useful indicator of aspect and one which doesn’t rely on the assumption of normality among the sampled elevations. The largest negative value of A is probably the best indicator of aspect. These functions can also be used for interpolation – predicting the elevations at in-between distances - but the reader is warned that linear least squares techniques can conceal a great deal of noise and so these functions must be used with caution when making a qualitative evaluation of aspect.
Overall, perhaps the best indicator of aspect is simply to evaluate the slopes themselves.
The feature is a currently existing site or object such as a town, a church, a ridge or hill, a sign, bridge, or other currently existing object or site which might be helpful in locating Bronze Age sites. Often it's the case that finding a currently existing feature is the best way to locate the BA site you're looking for. Features are displayed on site maps. Each feature has its own ID which is the letter 'F' followed by a three or four digit number. On a map they take the form of a red paddle with an upper-case F in the center.
In the following illustration you can see this feature/sign (arrow). If you mouse over the feature a box with the feature id and name will appear:
In this case it says 'F2073: Sign: Archaeological Site and Museum of Mycenae'
If you click on that icon an info-box will pop up that will give you more information about that feature.
Here we see that an infobox has popped up which gives the ID, the name, the lat/lon pair, the type, and its region. Also there is a link on the infobox. If you click on that link you'll be taken to a page created just for that feature.
On this new page the feature you clicked on, F2073, is now centered on the map and made larger than the other feature icons.
As I mentioned, each feature has a unique ID which is the letter 'F' followed by a three or four digit number. If you know the feature id that you want you can just type it into a search box. When you do that you'll be taken to a search results page which will have a link to your desired feature. So, for example, if you type 'F2073' (no quotes) in a Search box you'll be taken to a Search Results page which looks like this:
Here you see that there's a link to your desired feature, F2073, which turns out to be a sign to the Treasury of Atreus at Mycenae.
I took this (not very good) picture from Google Earth's Street View which shows the sign in question. Here we're facing SSW.Following this there is a working search box. Try it out!
Every map screen has buttons to support the generation of Comma-separated files (.csv) or Keyhole Markup Language (.kml) files. The philosophy I've adopted is this: the Mycenaean Atlas Project does not write to your hard drive. Instead a separate page is popped up with the .csv or .kml text on it. When you see it you can select it (Ctrl-A), copy it (Ctrl-C), paste it to an open editor window (Ctrl-V), and save this new file with the appropriate extension (.csv or .kml).
The comma-separated file (.csv) is a database file in which the individual fields are separated by commas. Such files can be cleanly imported, (not opened), into Google Earth.
A Keyhole Markup Language (.kml) file can be opened in Google Earth. In fact when you see a .kml icon on your desktop you can double click on it and it will open Google Earth (if GE has been installed).