Movement of sun - Palau (Jellyfish lake)
>> the connection is limited to constructions, this is a natural phenomenon

Cavalries - Potsdam
>> the connection is about calvaries (http://en.wikipedia.org/wiki/Calvary)

Still on T-list - Dingvellir, Jelling, Birka, Chan Chan, Cuzco, Machu Picchu, Cuenca, Quebrada
>> these are re-nominations as completely different WHS

Could have a new Connection??

On T List for a different nomination
Inscribed WHS (or parts thereof) which are on the T List for an additional (different) inscription from that already achieved

e.g Þingvellir Viking Monuments and sites http://whc.unesco.org/en/tentativelists/5587/

etc etc


Those sites would presumably get Connected as "Inscribed twice or more" if and when they get inscribed.

Re still on T list
I thought this was exactly what this connection was about. le corbusier for example is also acompletely new nomination according to different criteria.

Yes a confusion has certainly crept into this Connection

Originally it was intended to highlight WHS which were still on the T list for what seemed like the original inscription reasons - either because of administrative "incompetence" or from genuine "double counting". I remember that one of the "originals" was Gambia's Fort Bullen which had been inscribed as part of James Island etc but was also still on Gambia's T List in its own right! Siince then, Gambia must have cleaned up its T List and doesn't currently have one at all so Fort Bullen is only registered on this Web site now under "Sites that have been Nominated in the Past " with my and Ian Cade's reviews of it - http://www.worldheritagesite.org/sites/t245.html - and has been "removed" from this Connection.

Of those sites currently within the "Already inscribed, still on T List" connection, there seem to me to be 2 "incorrect" ones
a. La Chaux de Fonds/La Locle. This includes a Le Corbusier building but was of course NOT inscribed for that reason - therefore the inclusion of that building in T List entry for Le Corbusier buildings is NOT for the same nomination and it should be in my suggested new connection "On T List for a different nomination" ("Inscribed WHS (or parts thereof) which are on the T List for an additional (different) inscription from that already achieved")
b. Spissky Hrad This belongs to the "Extensions on T List" connection which is intended to highlight for us those WHS which have an extension planned on the T List.

To recap. There are genuinely (at least?) 3 different situations concerning inscribed sites and T List entries which justify separate "Connections"
a. Inscribed but with an extension for the same site on original reasons (or nearly original - no doubt there can be e.g an additional Criterion envisaged but in this case the entire original site would be so altered). Presumably many of the extended WHS once had a T List entry of this type before the extension got inscribed (Though, on reflection, I am not entirely sure if extensions are only allowed if they were previously registered on the T List!).
b. Inscribed but still on the T List either as the same site or as a part of the existing site (E.g one of the Troodos churches WHS also being on the T List in its own right) -this is the only one of the 3 situations whereby the T List entry couldn't, by definition, get inscribed
c. Inscribed but on the T List as part of an intended completely different site. This situation would only seem to occur where there is to be a serial site which uses an existing site or part of an existing site for different reasons and, if inscribed, would leave the existing site as a separate inscription with its own boundaries and critieria - a genuine "Inscribed twice" as per e.g the Belfries and Bruges.

(4488) Tidal effects
high coast
>> please explain, I only notice no or little tidal effects in the AB evaluation

Tides in the Baltic are VERY small, are not visible in the "normal" sense of the word (though a tidal flow from/to N Sea/Atlantic does occur and any body of water will itself be directly impacted by the pull of moon and sun) and any tidal level changes are overwhelmed by short term meteorological factors such as wind surging.

The Nomination File specifically refers to the "lack of tide" being a feature common to both High Coast and Kvarken. Later it states "The tide is less than 10 cm. However, winds and changes in the air-pressure make the sea water level highly variable, in some cases as much as 250 cm (Svensson 2002). Low water levels are common in spring, while higher water levels are common in the autumn".

I can only assume that the suggestion relates to the changes in water level resulting from Isostatic Rebound?

I've added this connection now

(4479) Historical Graffiti
rock islands
>> I found this quote in the AB ev, but is it Pre-20th century?
"The major rock art site on Ulong
Island is partly destroyed by graffiti. Incised or painted
directly onto the prehistoric red-painted art, the graffiti
consists mainly of the names and dates of visitors."

2. (4477) Traditional Hunting
rock islands
?? source?

3. (4475) Stepwells
rock islands
??source??

3. (4459) Pygmy Peoples
Rock islands
??source??

4456) Cocos also belongs to formerly inhabited
>> please name source, scientists alone doesn't count

Cocos IS (correctly) connected to "Formerly Inhabited" - what is the issue?

you're right, I did overlook that it was already there

what I meant was that each formerly inhabited island is by definition uninhabited. Perhaps we shoud better draw a distinction here. Maybe never inhabited?

The Boolean combinations for connections in this "domain" are quite complicated!
Potentially we have
a. Currently inhabited islands – Included just for logical consistency but not of interest for a connection - there would just be too many!
b. Currently uninhabited islands – Covered by the connection "Uninhabited Islands"
c. Currently uninhabited islands which were once inhabited – Covered by the connection "Formerly Inhabited Islands"
d. Currently uninhabited islands which were never inhabited – Not currently an identified Connection

Each of these combinations is then potentially increased by including the requirement that either "all" or else merely "some" of the islands in the site must possess the required attributes!! And, as an additional complexitiy, c and d above have 2 different attributes (e.g "Currently uninhabited" AND "Once inhabited") which could each be considered as applying either to the entire site or to part of the site! On this matter we have currently adopted different positions for b and c where Phoenix Islands are included in c when "Only 1 of the 8 islands is currently inhabited" but all the b sites are totally uninhabited.

On the other hand....... If a site is now totally uninhabited but was once partly inhabited is that not of interest (rather than the converse where an island is still partly inhabited but was once "more inhabited" - i.e The Phoenix Islands situation!)? I note for instance that Gough/Inaccessible COULD be added to "Formerly Inhabited". Wiki states for Inaccessible Island "The Stoltenhoff brothers, who arrived on Inaccessible from Germany in 1871, lived there for several years intending to make a living sealing and selling their wares to passing traders (forgetting how infrequently Inaccessible had visitors). However, due to the scarcity of food, they were "overjoyed" to be rescued in 1873 during HMS Challenger's visit to examine the flora and fauna there". Gough on the other hand has NEVER been inhabited.

I am not sure - we don't want to knock out potentially interesting connections. 10 minutes ago when I started writing this I was in favour of ONLY including "complete" sites across all atributes but now perhaps favour allowing some partial sites. But we need a definition which excludes eg 1 island out of 100 being "formerly inhabited" etc!

I suggest that
1. The "Uninhabited" (b) connection must apply to the entire site
2. The "Formerly Inhabited" (c) connection should apply only where the entire site is currently uninhabited but formerly at least some parts were inhabited. This would allow in Gough/Inaccessible but not Phoenix as I regard the former situation (ie "not now inhabited at all but formerly partly inhabited") as being of more interest than the latter ("still partly inhabited but formerly more inhabited")! It is a "judgement call" - what do others think?
3. Any new (see below) "Never Inhabited" (d) connection should apply to the entire site for both required attributes

Clearly there is a big overlap between b and c such that all sites connected on b must also meet c (but not necessarily vice versa). As a result "Bikini Atoll" should be added to "Uninhabited Islands".

I guess we could introduce a further connection for d ("Complete Island sites which have never been inhabited"). It would by definition be "b minus c". There could be a slight problem in knowing whether an island has NEVER been inhabited - can we really be sure about "never"? The way round this would be to adopt a definition which is really the reciprocal of that for c. i.e "No written records or human archaeological remains have been discovered"

Of those sites in b - only Surtsey, Wrangel and Heard/McDonald would meet all the criteria for this new connection. I have found this on Wiki regarding Auckland Islands (part of NZ Sub Antarcic Islands) "Traces of Polynesian settlement, possibly dating to the 13th century, have been found by archaeologists on Enderby Island. This is the most southerly settlement by Polynesians yet known.". As a result NZ Sub antarctic should be added to Formerly Inhabited (if we agree to adopt the "Not now inhabited at all but formerly partially inhabited" formula!)

I've now changed the Uninhabited and Formerly inhabited as per above.
I don't see much in adding another option (D - Never inhabited), as it is the obvious C - B, and I don't think there's anything interesting to tell about it (more than already has been done at the description of the site at B).

So applying to our "Connection rules":
a. not be "self evident"
...
d. add some knowledge or insight (whether significant or trivial!) about WHS for the users of this site

Could add NZ Sub-antarctic Islands to "Polynesia" on the basis of "Traces of Polynesian settlement, possibly dating to the 13th century, have been found by archaeologists on Enderby Island. This is the most southerly settlement by Polynesians yet known.".????

We have Polynesia as a connection under "Geography" rather than as a "cultural" connection but, rather than introduce an additional one for "Polynesian Culture" (which would almost duplicate the Geogrpahy one) we could perhaps use it with an definition along the lines of

"WHS within the geographical area defined as "Polynesia" and/or including Polynesian archaeological remains"

It was interesting to see the addition the other day of "Greater Blue Mountains" to the "Situated on a Continental divide" Connection – it is amazing how long perfectly "obvious" connections can remain unidentified!!

I note that it is linked on the basis that the Murray/Darling basin flows into the Indian Ocean. This is correct as per IHO definitions (and hence our own) which regard the Great Australian Bight as an arm of that Ocean. However, if you speak to Australians, buy Australian maps or read Australian articles they are most likely to regard the Bight as part of the "Southern Ocean"! This latter "Ocean" was only carved out of the Indian/Atlantic and Pacific Oceans by IHO in 2000 and its boundaries were set at 60 south (although there were various views both about this boundary and the ocean's name – Argentina didn't even want it created!) See here for the history of its creation and a resumee of different views regarding its boundary http://en.wikipedia.org/wiki/Southern_Ocean

There is however a second issue about our definition. When I suggested the Connection a couple of years ago I adopted a "hardline" definition of "Continental Divide" based on the requirement for the rivers across such a Divide to flow into different Oceans – with Oceans defined such that all open waters (excluding e.g Caspian and Aral) are part of a connected "World Ocean" which is in turn divided into 4 (or 5 with the Southern Ocean) and all "seas" are considered part of one or other of these oceans. Thus the Gulf of Mexico AND the Mediterranean Sea are both part of the Atlantic Ocean.

This definition has perfectly reputable provenance but isn't the only possible one – another allows them to flow into "different Oceans OR Seas". See this article using North America for examples. https://www.dmr.nd.gov/ndgs/NEWSLETTER/NLS03/pdf/Divide.pdf

Based on our current definition neither Pyrenees –Mont Perdu nor Swiss Alps Jungfrau-Aletsch straddle a "Continental Divide" since the Mediterranean Sea is regharded as "part" of the Atlantic Ocean

We need either to remove these 2 or alter the definition. It probably wouldn't make a great deal of difference if we loosened the definition to the Wiki one of "A continental divide is a drainage divide on a continent such that the drainage basin on one side of the divide feeds into one ocean or sea, and the basin on the other side either feeds into a different ocean or sea, or else is endoorheic, not connected to the open sea."
We would need to add the Great Smoky Mountains NP which sits astride the Eastern Divide.
There could still be issues about Seas. Do the Baltic, Black and Mediterranean Seas each have a "divide" or is the Baltic (like the North Sea) regarded as part of the Atlantic and the Black Sea regarded as part of the Mediterranean! I can't think of any WHS where this might be an issue - but i may be wrong!!