If we assume that all races today evolved out of where they settled (eskimos in the cold, dark-skinned people near the equator), what was the original human? Were they dark, then faded?
The generally accepted theory is the “Out of Africa” theory, which says that homo sapiens arose in Africa. I’m assuming that they would have had dark skin, as the extra melanin is supposed to protect against sunburn. Then I suppose that as people migrated to other, less sunshine-intensive continents, their skin color would have “faded”.
BTW, they don’t like to be called “Eskimo” anymore, which is a derogatory word in an alien Indian language (to them) meaning “eaters of raw meat”. They prefer “Inuit”. 
What DDG said regarding the first paragraph. The “fading” helps because Vitamin D is synthesized as a result of sunlight striking the skin. Melanin seems to inhibit that activity, so groups that live where clothing covers more of the body more often have tended to “lose” the amount of melanin that will prevent Vitamin D production.
Regarding DDG’s second paragraph: it isn’t that cut and dried. For the most part, the arctic dwellers of eastern North America prefer the term Inuit, however, many groups in the west prefer Eskimo.
This has been debunked:
http://www.landfield.com/faqs/alt-usage-english-faq/
BTW, the first race probably did have dark skin. According to Edgar Rice Burroughs, anyway, the First Born race was indeed black.
I can confirm the eskimo thing. I’m from Alaska, and every eskimo I ever met refered to themselves as “eskimo”. I know in Canada many refer to themselves as “Inuit”, but in AK they don’t. Maybe this will change, and “Inuit” will start to be the prefered word. But as of today, Alaskan eskimos still use eskimo.
But back to the OP. It seems very very very unlikely that the “races” evolved in place separately from Homo erectus or archaic Homo sapiens. We’re just too similar. How could each separate population evolve at exactly the same rate and in exactly the same way? No, modern races are almost surely very recent, less than 100,000 years old. But minor things like skin color and hair texture can evolve very quickly so we shouldn’t be surprised.
Lemur, have you been reading about the DNA analysis of the old Australian Mungo Man?
Mungo Man does not seem to have left any descedents, so his DNA is rather irrelevant to any discussion of current human diversity.
It is interesting how quickly his people reached Australia (if the dating is comfirmed) but at 60,000 years old they still had 10-20,000 years to walk out from Africa (from estimated date of “mitochrondrial Eve”). Plenty of time, even for government work.
I recall seeing somewhere that the Bushmen of the Kalihari are genetically different from all of their neighbors, and were thought at one time to be one of the extant oldest races, but new genetic analysis probably brings a different light to that contention.
BTW, Aleuts don’t like to be called Eskimo, or Inuit.
Hold on just a second, tomndebb. Evolution happens because of a positive genetic mutated trait. One that would ‘weed out’ the current norm. I don’t hear about a lot of blacks keeling over in, say, NY because they aren’t getting enough vitamin D. I don’t believe the human lifespan at that time was long enough for this to matter. And why the whole body and not just the covered parts. The earth is full of multi-colored lifeforms.
Ahem.
According to the Alaska Native Language Center,
Okay, it’s beside the point whether or not it means “eater of raw meat”, and exactly which people one should refer to as “Inuit”.
But just because some Native Americans from the Arctic may accept the use of the word “Eskimo” doesn’t mean they like it. Similarly, just because some African-Americans may accept the use of the word “Negro” doesn’t mean they like it, either.
So there. 
From How Much Vitamin D is Enough?"
italics mine
The same article mentions the prevalence of rickets in industrial cities prior to the increased supply of (Vitamin D rich) milk for children.
Evolution does not require that all changes have a goal. There are a number of peoples who live in temperate zones (and some near the arctic) who have darker skin than Europeans. However, a genetic change that resulted in less melanin would not be as likely to be selected against in the northern climes because the reduced melanin would allow better Vitamin D production. Such a change might propagate because it accompanied other traits that specifically increased chances of survival.
Why would only clothed parts of the body lose melanin? Do the genes know when we wear clothes? If a mutation occurred to reduce melanin production, it would tend to occur across the whole of the skin unless a specific feedback could be provided to encourage melanin production to remain high on the face, neck, and hands.
This article mentions the Vitamin D hypothesis–as well as objections to that hypothesis:
Human Pigmentation, U. of Indianapolis
DDG, As I recall, Canada and Greenland (actually Kalalit Nunaat, “Greenland” being a Norse real estate scam) are, indeed, east and Alaska west. 
Doc, don’t you think Mungo Man is an anatomically modern Hss?
That’s what he seems to be, but as I said he doesn’t seem to have left any descendants (the aborigines migrated to Oz well after his time) and there was still several thousand years between the earliest H. sap sap and his death in Australia. Plenty of time for his people to have moved from Africa and died off.