One of the reasons Cordain is unsure about saturated fats is because of a precontact frozen eskimo body that had heavy deposits in the arteries.  I came across a theory that might explain this, and tie together a few things.  The article is on Linus Pauling's theory of cardiovascular disease.  There's a summary here:

http://66.90.73.49/articles/show/222743-Pauling-s-last-legacy-a-unified-theory-of-cardiovascular-disease

Peer reviewed article here:

http://www.orthomolecular.org/library/jom/1992/pdf/1992-v07n01-p005.pdf

Vitamin C is needed for the synthesis of connective proteins such as collagen, which are important components of blood vessel walls.  Pauling posits that atherosclerosis is, essentially, a mechanism to maintain blood vessel integrity when vitamin C levels are inadequate to keep the blood vessel walls healthy.  The human inability to synthesize vitamin C, along with unreliable dietary sources of vitamin C, would then have created a strong evolutionary pressure for reacting to low vitamin C levels with atherosclerosis.  Pauling thinks the vitamin C intake became lower and less reliable as we moved out of the tropics, but given the evidence of meat eating from the early tropical African paleolithic, it seems likely that this actually took place earlier, as we shifted to meat as our primary food at the beginning of the paleolithic.

Pauling's theory is consistent with the eskimo data, because eskimos would have had limited access to fruit and thus little vitamin C intake.  According to Pauling, arterial deposits should normally increase during winter, when there was less vitamin C available, and decrease during summer.  Eskimos would have had very little vitamin C available even in summer, and so should be expected to accumulate the deposits.

Of course, neolithic foods which contained carbohydrate without vitamin C would have made things worse, because glucose is chemically similar to and competes with vitamin C.  I've previously noted the resulting negative effect on the immune system; evidently this applies to connective tissue as well.  Neolithic carbohydrate, by decreasing vitamin C, would both increase the chances of scurvy, and also increase susceptibility to atherosclerosis.

And of course, even with arterial deposits, a thrombosis is needed to cause a heart attack; neolithic foods tend to be inflammatory, increasing the probability of such a thrombosis.

I don't regard Pauling's idea as proven, but it does make a lot of sense to me.  There has also been some recent support:

http://onlinelibrary.wiley.com/doi/10.1111/j.1742-7843.2009.00420.x/full

My bottom line is that I'm going to be more regular about squeezing lemon juice into my water.

There's about 20mg of vitamin C per lemon.  The real question is how much we need.  I suspect that varies extremely widely depending on glucose intake from starch and sugar.

Pauling contended that our vitamin C intake should be about 5% of our glucose intake, because that's roughly the amount that gets converted in animals that can make their own vitamin C.  That's 10-20g, or 10,000-20,000 mg, per day for people on a starch heavy western diet.  On a typical paleo diet, it would probably be a couple of grams.  However, I have to believe that the adaptations we've made, including the ones Pauling has identified, would have made our requirements lower than that.  How much lower is a tough question.

Another source we could look at is the amount of vitamin C in wild fruit.  Here's a table that has some information:

http://www.naturalhub.com/natural_food_guide_fruit_vitamin_c.htm

Some fruit, such as the acerola and camu camu, are extremely high in vitamin C - a couple percent of their total weight, and likely 10% or more of their glucose content.  However, these fruit are from the new world and are likely adapted to the needs of frugivore monkeys, who haven't had to adapt to lower levels of vitamin C as humans have.  Baobab fruit might be a better example of what our paleolithic ancestors might have eaten, and its vitamin C content is between 1% and 5% of its glucose, based on the above link and here:

http://www.baobabfoods.com/faq

That might indicate that humans might be able to get by on vitamin C intake as low as 1% of glucose intake.

The only supermarket fruit that seem to have vitamin C content as high as 1% of glucose content are strawberries, kiwi fruit, and citrus, with lemons at around 2%.  That suggests that if one is eating other fruit with less vitamin C - for example apples - one might want to consider supplementation.

Of course, the other issue is that it's known that scurvy is prevented by meat, which doesn't seem to contain vitamin C.  I don't think we know why, though one might speculate that we might not need vitamin C to convert certain amino acids to their hydroxylated forms since meat contains connective tissue which already has the hydroxylated amino acids.

Anyway, I plan to start squeezing a wedge or two of lemon into my water each day even when I'm not having any carbohydrate.  If I eat much fruit that is low in vitamin C, I may add some vitamin C supplement as well.