On the Beach

 I am happy to announce that the users of the Mycenaean Atlas now have the ability to examine distances from Bronze Age sites to the closest shoreline.

The main control page now features a new way for the user to specify a low and high distance. 

The new control is in the lower center.

Here is the new control in close-up:

 Having specified the range the user will see a new page that displays distances in four ways: a map of the returned sites, a list of the sites with distances to the shore, a histogram in which the sites are distributed into 'same distance' buckets, and a line chart in which the sites are lined up in distance order.

Analysis of sites within 1000 m of the shore.

For example you may be interested in all sites within 1000 m of the shore.  You specify 0 as the minimum distance, 1000 as the maximum distance and then you click on submit.  The resulting four-part display will show you all sites within 1000 m of the closest shoreline: on a map, in a list, in a histogram, and on a line chart.

It may not be optimal to display the entire DB in this way; more researchers are interested in examining one island or region at a time.  The control allows for that.  If you want only to examine the Cyclades then you will type 'Cyclades' into the text input labelled 'Region' (and specify a range).  

When you click submit you will see that the display is limited just to the Cyclades.  

Result of selecting just the 'Cyclades'.

This atlas considers individual islands as 'subregions'.  If you wished to look just at Naxos in the Cyclades then type 'Cyclades' for the region and 'Naxos' for the subregion.

Selecting 'Cyclades' and 'Naxos' from the control

The analysis page for Naxos.

---

The individual place key site detail page has also been modified.  Now if a distance to the ahore is defined for the site then there will be a label on that page that says 'Distance to Shore (click)'.  When you click on that label a map will pop up that shows the site with a line drawn to the nearest shoreline.  Click on the map to pop it back down.

Click on this to see site to shore map.

The site to shore map with connecting line.  Zoomable.
Click on map to pop it down.

The concept of shoreline poses some problems.  'Shoreline' is a dynamic concept and no precision is ever possible.  There are the tides, of course.  Distance to nearest shoreline depends on the time of day.  And while I give figures precise to the decimeter the real distance may fluctuate tens of meters depending on the time of day.  Here in the United States we have a joke about the State of Rhode Island which is said to be '5% larger at low tide.' There have also been dramatic changes in the shoreline since the Bronze Age.  There is the case of Tiryns, of course.  Formerly on the water's edge it is now more than a kilometer inland.  Rivers have done their usual work in the last 3500 years.  What were formerly islands are now landlocked by deltas as is the case for Aetolia-Acharnania.  And there have been changes even more dramatic.  The city of Helike in Achaea was drowned by earthquake and land subsidence in 373 BC.  There have been changes in the ocean level which, combined with land subsidence, have led to differential variations in shorelines across the E. Aegean world.  Nisakoula in Messenia is an illustrative case.  Once a thriving settlement it is now submerged.  Likewise Paulopetri in Laconia.  The filling in of the Cothon harbor near Nestor's Palace in Messenia has placed that Palace further from the ocean than it would have been in the tthieteenth century BC.  So the accuracy of my figures can certainly be questioned and in very few cases are they even accurate to within 10 m.

  So if I cannot achieve absolute accuracy with my distances why, then, have I provided them?  My readers should consider the 'distance to shoreline' as a classifying device.  This new ability will quickly allow users to create lists of sites 'on the shore' (0 to 1000 m, say), 'a moderate distance from the shore' (2 to 5 km?), and 'inland sites' (perhaps sites further from the shore than 6 or 7 km).  It would be very difficult to generate these classes in any other way.  Why should my users care about these classes?  Well, for one thing, distance from the coast might very well be reflected in diet and occupation.  And there are obvious effects on the type of work that communities might perform.

How was this done and how were these figures arrived at?  The first step in such work is to create a new DB table just to store site identifiers and distances.  Populating such a table is the next step.  This consisted of drawing in Google Earth an outline of all the areas involved.  Such an outline consists of nodes and vertices.   This is time-consuming and I experimented with different approaches.  For Crete I traced the entire shoreline - some 4000+ nodes.  For areas such as Attica I used a circle and node technique.  In this method I drew circles centered on the various sites and only modeled those stretches of coast where the circles touched the shoreline.  This resulted in significant savings in both time and labor

I then wrote a .php routine that would calculate the distances from each site to ALL the nodes and save the shortest distance along with the corresponding lat/lon pair for that point/node on the shoreline.  In this system, as implemented, there is one and only one 'closest point' for each site even though many sites have access to the shoreline at two or more points at nearly identical distances.  Some of these access points that are a little further away are objectively better for ocean access than the closest point.  I save these difficulties for later discussion.  In a potential refinement the current table could be combined with the data from Arthur de Graauw's harbor data.

This work of outline tracing is not yet completed.  The areas that are done are

• Achaea
• Arcadia
• Argolid
• Attica
• Boeotia
• Corinthia
• Elis
• Laconia
• Macedonia
• Megaris
• Messenia
• Thessaly

Among the islands the work is completed for

• Antiparos
• Crete
• Euboea
• Kasos
• Kephalonia
• Kos
• Lesbos
• Melos
• Naxos
• Paros
• Rhodes
• Syros
• Thasos
• Thera

Segment Ro of the Cyclopean Wall is located again

  The segment of the Cyclopean wall which was named 'Ro' by Broneer (Mycenaean Atlas Project: C7760) has been relocated by my colleague, Peter Barkevics. Its coordinates are 37.913475° N, 23.003006° E. It sits at an elevation of 27 m and is about 125 m. in a straight line on a bearing, from St, of ~273.1°. It sits on an open hillside about 20 m. above a large private home with a gray ('blue' in the photo) roof. Here is a picture of the situation.

Hillside with Ro.  North at the top. Red line is the Hexamilion wall.
Google Earth Image.

Next is Barkevics' photograph of Ro looking directly east and downhill towards the house with the gray roof (which appears to be blue under open sky light) (37.913466° N, 23.003295° E).

Section Ro of the Cyclopean wall. 
Facing E and downhill.
Peter Barkevics.  All rights reserved.

In this picture the view is directly E; the house with the blue roof is clearly seen. On the left hand side can be seen the four projecting stones of Broneer's 'tower'. The other flanking stones are to the right. Compare this photograph to Broneer's drawing:

Drawing from Broneer [1966] 350, Fig. 2, no. 4.

In the next photo we see these stones from the other direction (looking west):

Section Ro of the Cyclopean wall. 
Facing W and uphill.
Peter Barkevics.  All rights reserved.

In this photograph the four projecting stones are now on the right. The large anchor stone visible in Broneer's drawing is at the lower left.

 That anchor stone is also clear in this next photo which looks at the east end of this segment while facing south.  Here the two east-side projecting stones (the 'tower') are on the right and stretching out towards the viewer.

East end of section Ro of the Cyclopean Wall. 
Viewer facing S.
Peter Barkevics.  All rights reserved.

Bibliography

Broneer [1966] :  Broneer, Oscar. ‘The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology’, Hesperia: The Journal of the American School of Classical Studies at Athens (35:4). 1966.  Online here.

Broneer [1968] : Broneer, Oscar. 'The Cyclopean Wall on the Isthmus of Corinth, Addendum', Hesperia: The Journal of the American School of Classical Studies at Athens (37:1), 25-35.  The American School of Classical Studies at Athens. 1968.

Gregory [1993] : Gregory, Timothy E., Isthmia V. The Hexamilion and the Fortress, American School of Classical Studies at Athens. Princeton, New Jersey. 1993. ISBN: 0-87661-935-9.

Kardara [1971] : Kardara, Chrisoula. 'The Isthmian Wall; (A Retaining Wall for a Road)', Athens Annals of Archaeology (4:1), 85-89. 1971.  Online here.

Morgan [1999] :  Morgan, Catherine. Isthmia VIII; The Late Bronze Age Settlement and Early Iron Age Sanctuary. The American School of Classical Studies at Athens, Princeton, New Jersey. 1999.

Simpson and Hagel [2006]:  Simpson, R. Hope and D.K. Hagel. Mycenaean Fortifications, Highways, Dams and Canals. SIMA vol. 133. Paul Åströms Förlag. Sävedalen, Sweden. 2006.

Photographs of section Pe of the Cyclopean Wall

 I mentioned in a previous post that my associate, Mr. Peter Barkevics, has confirmed the location of segment Pe of the 'Cyclopean' wall near Isthmia in Greece.  The position of this segment is 37.913457° N, 22.997472° E.

Fig. 1.  Area of the Isthmus.  Blue way-mark
indicates section Pe of the 'Cyclopean' wall.

Zooming in:

Fig. 2.  Area of the 'Cyclopean' wall.  Section Pe is in the center
at the blue way-mark.

Section Pe was reported by Oscar Broneer in 1966 [1] and he provided a drawing of it:[2]

Fig. 3.  Broneer's drawing of section Pe of the 'Cyclopean' wall.

Overlaying the site re-established by Barkevics with Broneer's drawing gives this:

Fig. 3a.  Google Earth image overlaid with Broneer's sketch of Pe.  Broneer [1966] 350, fig. 2, no. 3.  Elevation of the wall at the way-mark is 60 m. a.s.l.

This wall segment is about 6 m from the edge of the modern road.

Broneer described Pe like this: 

"One continuous stretch, Pe (Fig. 2,3; Pl. 81), in the property of Sotiris Peras, measures 45.50 m. in length. Only the outer, northern face is well preserved, in places to a height of two courses, but a few stones from the inner face show that the wall here had a thickness of 3.60-4.00 m. The stones are large, some measuring 1.50-1.75 m. in length and 0.75 m. in height.  There are four towers, varying in width between 2.10 and 2.60 and projecting ca. 0.70 m. from the face of the wall. The distance from one tower to the next varies between 7.90 and 9.50 m.  There was very little earth close to the wall, and the sherds found on the surface are mostly small pieces of undatable coarse fabrics. Two undecorated pieces might be Mycenaean."[3]

Two previous photographs of Pe are known to me - one appeared in Broneer [1966] [4].  The other appeared in Morgan [1999]. [5]  To these old and now outdated photographs Mr. Barkevics has added some fifty color photos which detail the segment stone by stone.  At the bottom of this next figure we see Mr. Barkevics diagram notes of the positions where his pictures were taken and, above that, my own diagram that tries to match his photos to Broneer's drawing.

Fig. 4.  Upper drawing is Broneer [1966] 350, Fig. 2, no. 3.
Lower drawing is Google Earth overlaid with Barkevics' way markers.

Is this segment photographed by Barkevics the same as that shown in Broneer's photo?  Yes.  This next photo shows my labels for several of the stones in the Broneer photo:

Fig. 5.  Broneer [1966] Pl. 81.

This next photo by Barkevics shows the same stones with corresponding labels:

Fig. 6.  Tower 3 (from W).  View is to S.

I believe that stone beta has been moved and I may not have identified it correctly.  Stone beta may actually be the next small stone (with the question mark) to the viewer's right.  The other stones are clearly the same.

Following is a tour of the structure from the west to the east.

Fig. 7.  Photo 325.  The first two stones of the N face on the west. 
Viewer is looking S.

This photo shows the first two stones in the sequence, starting from the west.  This is photo 325 and you can see its position by looking at the left side of fig. 4 in this post.  Here are the same two stones again (note the mossy patch).

Fig. 8.  Photo 326.  West end of N face.  View facing S. 
The person is standing on the north edge of the modern road.

Fig. 9.  Photo 327.  First tower from the W.

This is the first (counting from the W) of the projections which Broneer identified as 'towers'.  Their small size and the nature of the ground strongly suggest that they are buttresses.  

The next photo shows the view from the N of this same 'tower'.

Fig. 10.  Photo 328.  Same tower from the N and looking S.

The second 'tower' from the left is shown in the next photo:

Fig. 11.  Photo 332.  Tower 2 (counting from the W).

... and another view of Tower 2:

Fig. 12.  Photo 333.  Tower 2 from a bit further back.

The next photo gives another view of Tower 3 (already shown in Fig. 6):

Fig. 13.  Photo 346.  Tower 3.  The view is to SE. 
Ten-euro note (127 x 67 mm) for scale.

Another view of Tower 3:

Fig. 14. Photo 348.  Tower 3 from N. 
View is directly S.  Compare to Fig. 5.

Fig. 15.  Photo 351.  East end of Tower 3 (from W).

Fig. 16. Photo 357.  Tower 4 (counting from the W).  View from the E.
Underbrush suppressed for clarity.

It is difficult to escape the impression that section Pe is slowly being degraded by natural forces.  Both erosion and the undergrowth (particularly pine trees) are going to dislodge more stones.  The Archaeological Service might do a great deal of good by spending a day clearing brush around this very important monument.

In future blog posts I will publish more of Barkevics photos and then try to say what we have learned about the Cyclopean wall.

Footnotes

[1] Broneer [1966] 351.

[2] Broneer [1966] 350, fig. 2, no. 3.

[3] Broneer [1966] 351

[4] Broneer [1966] Plate 81.

[5] Morgan [1999] 443, fig. 8.  Mislabelled 'Section Sp'.  The correction is due to Simpson and Hagel [2006] 130, fn. 103.

Bibliography

Broneer [1966] :  Broneer, Oscar. ‘The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology’, Hesperia: The Journal of the American School of Classical Studies at Athens (35:4). 1966.  Online here.

Broneer [1968] : Broneer, Oscar. 'The Cyclopean Wall on the Isthmus of Corinth, Addendum', Hesperia: The Journal of the American School of Classical Studies at Athens (37:1), 25-35.  The American School of Classical Studies at Athens. 1968.

Kardara [1971] : Kardara, Chrisoula. 'The Isthmian Wall; (A Retaining Wall for a Road)', Athens Annals of Archaeology (4:1), 85-89. 1971.  Online here.

Morgan [1999] :  Morgan, Catherine. Isthmia VIII; The Late Bronze Age Settlement and Early Iron Age Sanctuary. The American School of Classical Studies at Athens, Princeton, New Jersey. 1999.

Simpson and Hagel [2006]:  Simpson, R. Hope and D.K. Hagel. Mycenaean Fortifications, Highways, Dams and Canals. SIMA vol. 133. Paul Åströms Förlag. Sävedalen, Sweden. 2006.

Section St of the 'Cyclopean' Wall is found

 Peter Barkevics, has re-established the location of segment St of the 'Cyclopean Wall'. [1]  It is at 37.913404° N, 23.004431° W.  Formerly this was on an olive grove owned by the Statiris family.[2]  After being described by Broneer et al, some years ago, it was acquired by its present owners.  The wall is positioned in the front yard of a bed and breakfast.  Some stones appear to be occluded by building in the back of the property.  This section of the wall is 237 meters from segment Sk on a bearing, from Sk, of 282.79°.

The owner allowed Pete to photograph freely but would only allow two reduced-size images to be reproduced on this blog.

North face of St looking W

North face of St looking E

All photographs are the property of Peter Barkevics and all rights are reserved.

In this next photo we see the B&B property as it looks in Google Earth:

Here is a copy of the drawing that Broneer made of St.[3]

In this final photo I superimpose Broneer's drawing over a GoogleEarth image of the B&B in question.

Here I have fit the drawing to the stones as best I could. It seems clear that the north face runs E-W through the center of the lawn. The south face is partly covered by the driveway and, in the back, by the building of a shed.

Peter Barkevics has interesting observations about the current condition of this wall. He writes:

"The current owner, ..., says there have been several floods of water from Mytika area after heavy rain in past years. A culvert has been built between the houses above his and it discharges over his courtyard. The discharge has eroded some of the area and the wall was affected. ... All I can say is that a quick survey of the site by the archaeological service would be a good idea. It's only a day's work for a survey team ... "

I want to thank Peter for his untiring work in finding and documenting these wall fragments.  They are important for understanding Greece's Late Helladic history. Shamefully they have so far been given, not analytical, but mythologizing treatment.

Footnotes

[1] First described by Broneer and Kardara. For section St see Morgan [1999] 440, no. 2, 'Section St (Statiris Property)'. Also Broneer [1966] 349 and fn. 8.

[2] Broneer [1966] 349.

[3] Broneer [1966] 350, fig. 2, no. 5.

Bibliography

Broneer [1966] :  Broneer, Oscar. ‘The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology’, Hesperia: The Journal of the American School of Classical Studies at Athens (35:4). 1966. Online here.

Kardara [1971] : Kardara, Chrisoula. 'The Isthmian Wall; (A Retaining Wall for a Road)', Athens Annals of Archaeology (4:1), 85-89. 1971.  Online here.

Morgan [1999] :  Morgan, Catherine. Isthmia VIII; The Late Bronze Age Settlement and Early Iron Age Sanctuary . The American School of Classical Studies at Athens, Princeton, New Jersey. 1999.

Section Pe of the Cyclopean Wall is found

 The segment Pe (C7761) of the Cyclopean Wall has now been confirmed to be located at 37.913457° N, 22.997472° E. My associate in Greece, Mr. Peter Barkevics has now confirmed its location and provided a large collection of photographs. The elevation of this segment is 65 m a.s.l. It is located just below the road and about 18 m. (59 ft.) below the edge of the Mytika plateau and about 20 m (65 ft.) above the valley floor.

In this view we are facing the Saronic Gulf to the SE. The modern road runs almost perfectly E to W. The distance from the edge of the road to the blue paddle labeled ‘C7761 Section Pe’ is almost exactly 6 m. Here the wall has been sketched in with an orange outline; it has been superimposed on the drawing by Broneer [1966]. [1] The circle in 45 m. in diameter.

Drawing from Broneer [1966]

The segment is, of course, in a neglected state; heavily overgrown and concealed by brush. The characteristic ‘towers’ identified by Broneer make the identity of this wall unmistakable. Moreover the position agrees with the map in Frey and Broneer [2016]. [2]

I include some of Pete's great photos and I will publish more soon.

Here is a  photo (taken from the W) of the W edge of the second tower from the left in the Broneer drawing:

Here is a photo of the west corner of Tower 3 (from the W):

There are some dangers to be aware of.  The next photo shows an old excavation (?) trench which is just a little way to the E of Pe:

I will add more of Peter's photos to this blog post (so check back) as soon as they are post-processed and labeled.

For right now we should all thank Peter Barkevics who not only went out, found, and  photographed this wall.  He also identified the Frey and Gregory article which showed the right position of Pe.  

If it had been up to me we all would still be looking at the edge of the Mytika plateau.

Footnotes

[1] Broneer [1966] 350, Fig. 2, no. 3.  Comments in red are mine.

[2] Frey and Gregory [2016] 438, Fig. 1.

Bibliography

Broneer [1966] : Broneer, Oscar.  "The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology",  Hesperia: The Journal of the American School of Classical Studies at Athens (35:4), 346-362.  1966.  Online here.

Frey and Gregory [2016] : Frey, Jon M. and Timothy E. Gregory, 'Old Excavations, New Interpretations: The 2008–2013 Seasons of The Ohio State University Excavations at Isthmia', Hesperia: The Journal of the American School of Classical Studies at Athens (85:3) 437-490. 2016. Online here. 

 Where is Section Pe?

The picture of Pe that Broneer [1966] presents looks like this:[1]

In this picture arrow b shows a left-leaning (to S) branch on a young tree.   Arrow a points to one of the buttresses.

This next picture is labeled in Isthmia VIII as a section of Sp.  [2]

But Simpson and Hagel [3] say that this is mislabelled and is really a photo of Pe.  And that makes sense. The wall in this photo runs up a rather steep slope which we know Sp does not.  Also the buttresses seem to be visible at arrow a.  The right-leaning (to S) branch at arrow b seems to match the left-leaning branch on the other photo.  So if all this is correct Pe is not likely to be where I thought it is.  It's more likely to be further to the E (about 100 m) near 37.913481° N,  22.998581° E where there is a clump of trees.  It looks like this in Google Earth Street View (facing W): 

Here the arrow at 'a' point to a clump of trees in which the first and seconds pictures might have been taken.

This next picture shows what it looks like from the other side and facing E:

I think a candidate for Pe that would match the second photograph is somewhere just beneath the edge of this clump of trees.  It looks like there is a very convenient turnout right next to it.

From above the whole scene looks like this:

Here North is at the bottom.  'P1' and 'P2' in red are the locations of the two Street View scenes.  C7761 (Pe or where I was thinking that Pe might be) is in the center at the blue paddle.  The circle radius is 100 m.  The turn-out in the previous photo is just under the yellow push-pin over the group of trees at the left.

And, not to forget, this section may involve a fall hazard so be careful of your footing.

Footnotes

1. Broneer [1966] Plate 81.

2. Morgan [1999] 434, figure 8.  Labelled incorrectly 'Section Sp'.

3. Simpson and Hagel [2006] 130, fn. 103

Bibliography

Broneer [1966]:  Broneer, Oscar. ‘The Cyclopean Wall on the Isthmus of Corinth and Its Bearing on Late Bronze Age Chronology’, Hesperia: The Journal of the American School of Classical Studies at Athens (35:4)

Morgan [1999]:  Morgan, Catherine. Isthmia VIII; The Late Bronze Age Settlement and Early Iron Age Sanctuary . The American School of Classical Studies at Athens, Princeton, New Jersey,

Simpson and Hagel [2006]:  Simpson, R. Hope and D.K. Hagel. Mycenaean Fortifications, Highways, Dams and Canals. SIMA vol. 133. Paul Åströms Förlag. Sävedalen, Sweden. 2006.

Placement of C1542. The Skala Oropos Mound.

I have heard from a correspondent who visited Greece recently and who has this to say about my placement of C1542 which I called the 'Skala Oropos Mound'.

He says this:

“C1542 Skala Oropos: Mound

38.315918° N, 23.809185° E is not quite accurate. The mound is nearby, covered by a grove of olive trees and bounded by houses, at 38°18'56.6" N  23°48'26.8” E - if this is indeed the mound: I saw no sherds in the ploughed earth around the olive trees.”

I remember struggling over this. Basically there are two sources. There is the Simpson/Dickinson/Arch Reports/Phialon track. Then there is the Cosmopoulos (Oropos Survey Project) track.  Cosmopoulos' survey effort seems to have been done independently of Simpson et al.

What does Simpson [1981] say? At p. 52, ‘B45 Skala Oropou’ he says:

“Prehistoric sherds have been found on a low mound, about 100 m. in length, about a kilometre to east of Nea Palatia, a suburb of Skala Oropou. The site is described as lying near the edge of the coastal plain and about 500 m. from the sea, at the point where the road from Skala Oropou to Markopoulo cuts a disused mine railway track which runs part (sic) the north and east foot of the mound.”

The time I’ve spent looking for that ‘disused mine railway track’! Failing that Simpson (and Simpson/Dickinson [1979]) clearly refer to the hill you specify. No other hill even comes close. And the word ‘low’ might apply.  This hill rises about 20+ m. above the surrounding plain. My marker for C1542 was just off the hill to the E and it seems like moving that marker up to the shoulder of that hill might be appropriate. The length of the entire hill is more than 400 m. from W to E. The shoulder is 60-80 m. long.

Cosmopoulos [2016] refers to that hill as ‘Ψ’ and says this (p. 104, 91/9. Agios Nikolaos): “This is a large findspot in the long and narrow valley immediately to the west of hill Ψ (Figure 35).

The valley runs in a north-south direction and can be accessed by St. Nicholas Street, which is named after the small chapel in its southern end. A stream bed runs through the valley in a southwest-northeast direction and turns to the east where the alluvial plain starts. The valley and the hillslopes around it are covered with vineyards, olive groves, and fig trees.”

I reproduce Figure 35 below:

Reproduction of  Figure 35 in Cosmopoulos [2016] 57.
Phioto taken from approx. lower center with axis shown by the arrow.
C1542 is shown on top of the 'Ψ' hill at upper center.
Area of Cosmopoulos' finds (91/9) is approx. the oval.

The chapel of Hagios Nikolaos is here: 38.316245 N, 23.810375 E. Cosmopoulos is describing a very large (basically the entire valley floor) site just off the E side of hill Ψ. He reports no prehistoric finds from this area. All the finds were Classical and later. It does seem strange that he seems to know nothing about Simpson et al.

Action Taken. I moved the location of C1542 up onto the shoulder of the hill. My new suggested position is 38.315505° N, 23.808465° E. I have left the documentation alone. In this context (since this is a BA atlas) I have ignored Cosmopoulos.   

No changes will appear until I release a modified database.

Bibliography

Cosmopoulos [2016] : Cosmopoulos, Michael B., The Rural History of Ancient Greek City-States; The Oropos Survey Project.  BAR International Series, 1001.  2001. 

Simpson [1981] : Simpson, Richard Hope. Mycenaean Greece. Noyes Press.  New Jersey. 1981