I find the effects on soil to be very fascinating, especially in the cases where the soil has become 'baked', and is unable to retain water. When you have the two different types of soil side by side, the difference is visible and very striking. Let me assure you, I am not a scientist, but you really don't have to be to see the dramatic difference in the appearance of the soil. If anyone would like to see a great picture of the differences in the soils from a well researched case, if you get a copy of "Left At East Gate" by Peter Robbins and Larry Warren, in the photo section in the book is a picture of the soil samples that Peter collected from Capel Green, in the Rendlesham Forest. As you can see from the photo the differences are striking.
When you are faced with a case that has possible soil evidence, the best way to collect the soil samples is to get a bunch of 1 foot sections of PVC pipe, and end caps for both ends. What you will want to do is drive each PVC section into the ground at least 4-6 inches, starting at the center of your investigation grid in a straight line outward, driving them in about every 1-2 feet, depending on the size of your grid. You will want to get control samples as well, so you will want to extend your line an equidistance outside your grid so you have the same number of control samples.
Carefully remove each PVC section, capping both ends, and marking all the information on the section in Sharpie or marker. You will want to include the number of the sample, the date and time, who retrieved the sample, and the case number. Remember to log each sample in your evidence log, and to keep track of the chain of custody if you hand it over to anyone else.
If you follow these steps, you can rest assured that you will have a properly retrieved soil sample.