Experts recommend that you take your soil samples in the fall if you're likely to need to add lime (or, presumably, sulfur) to change the pH of your soil since the addition requires time to react.  Personally, I prefer to take soil samples in the winter because the ground is very easy to dig into at that time of year (and since our soil is already sweet).  But, in reality, the best time to take a soil sample is whenever you think about it and need some data.

Which is a long explanation of why I was filling a flower pot with dirt from several spots in our starplate pastures, mixing it up, and then tossing a representative sample in a ziploc bag to go in the mail.  Although the texture of the earth in that area is excellent, there's clearly a major deficiency at play since very few plants felt like growing over the summer.  When the primary trees in an area are black locust and sassafras and when even comfrey fails to thrive, you know you need a soil test.

Why did I feel the soil test was so critical that it couldn't wait a few more months?  I've been itching to add a ruminant to our farm because they can get most or all of their nutrition from pasture, but there's a flip side to that coin.  If your soil is deficient in a mineral and you expect animals to get all of their food from that plot of earth, they'll end up deficient in the same mineral.  Starved soil could mean starved goats, and I don't want to risk it.  So soil-testing in September it is.