There was a reference a while back in this thread. You remember it right? You "debunked" it for us. The reasons for alteration were stated, the raw, and altered plot was shown.
It's doctored. The argument is that it's supposed to have been doctered that way, but he shows the doctoring they did.
Yes I remember but it's pretty difficult to just find it all through the 24 pages of replies of this thread but OK I went back and checked out what I went over in response to
The Smoking Gun At Darwin Zero article by Willis Eschenbach.
In point of fact I posted 8 separate links to articles rebutting various aspects of Mr. Eschenbach's argument. Did you read them all? I've obviously gone over all of these before but I guess I'll have to do it again. I'll leave out a few of the links for brevity's sake.
One of the last ones was in fact a statistical comparison between the raw and adjusted GHCN/CRU datasets. This also happens to address Daiwa's question from reply #586 where he asks why all adjustments are only in the upwards direction. The answer is that they are *not*. It's just that deniers only point out positive adjustments and make no mention of the negative ones.
The following graph is from http://www.gilestro.tk/2009/lots-of-smoke-hardly-any-gun-do-climatologists-falsify-data/ which as I said compares the raw and adjusted GHCN/CRU datasets. Note the mean adjustment is 0 and the average adjustment is 0.017°C/decade which is less than 10% of the average warming trend over the last century of 0.2°C/decade. So even if all the adjustments were bogus it would only account for 10% of the reported warming. But they aren't bogus and there's data to prove it (see the next link).
This proves that adjustments (at least in the GHCN/CRU dataset) are made in both positive *and* negative directions and that the net total adjustment sums pretty close to zero.

Also here's a report from the Australian Bureau of Meteorology that goes into quite a bit of detail about the precise why and how these adjustments were made (at least for the Australian stations). Note that every single Australian station is listed along with its location, elevation, start year, quality rank, whether its urban or not, whether its included in the high quality dataset and it isn't the reason its not included.
http://reg.bom.gov.au/amm/docs/2004/dellamarta.pdf
Here's a similar document that also goes into quite a bit of detail regarding adjustments to individual Australian stations. Basically data from individual stations are examined for discontinuities that do not exist in nearby stations and statistically corrected. A representative record from a single station is given as to the magnitude and direction of adjustments as well as the reason. The point is that such data exists for *all* stations that have been adjusted.
http://134.178.63.141/amm/docs/1996/torok.pdf
Now here is one of the earlier rebuttals that I posted and admittedly you're correct in saying that all this one does is to make the argument that it's supposed to be that way however it did post a graph of the record from the high quality station at the Darwin airport and note that compares very closely to the "adjusted" data that seems to be causing such a fuss.
http://thingsbreak.wordpress.com/2009/12/09/the-smoking-gun-darwin-station-temperature-adjustments/
Here's another one that makes pretty much the same argument as the above article and posts essentially the same graph from the Darwin airport shown below. However it does make one additional point and that is that Mr. Eschenbach makes the mistake of treating measurements from different sites as if they came from the same site. Now I'm not a statistion so I'm not sure why this is a problem but apparently it's either simple incompetence or it's intentional fraud.
http://scienceblogs.com/deltoid/2009/12/willis_eschenbach_caught_lying.php

The bottom line is that the adjustments are both justified and documented, and even if they weren't would at most reduce warming by less than 10%.
There is no smoking gun. There is no evidence of foul play whatsoever. This is indeed debunked way more than should have been humanly necessary but you really do actually have to read the data presented to make any kind of honest decision.