Let's start with micron. Think of micron as a ruler. A ruler is divided into lengths 1/8, 1/4, 1/2, 3/4, 1 inch so you can measure different lengths. Micron is the measurement of thickness like you would measure the diameter of a circle. When applied to alpaca fiber it is the measure of the thickness of fiber but when reported it is the average thickness of many fibers in a sample. Relative to humans, most cannot distinguish variation under 3 micron with the eye or feel but generally can distinguish prickle over 25 micron. The lower the micron of a fleece the smaller diameter of the fibers and the less prickle and more comfort next to the skin.
So when we talk about micron average what is the range of the sample or put differently how many fibers were thicker than the average and how many were thinner than average.
That is where SD comes in. This is a statement of variability of the sample of the fleece. IF SD is 4 and the average micron (AFD) is 22 this tells you that half the fibers below the average were between 18 and 22 micron and half the fibers above the average were 22-26 micron. So what SD is telling you is how close to the average is the sample.
CV is calculated by dividing SD by AFD. So in our example CV has little bearing or by itself little relevance. Where CV comes in handy is when you have a bale of fleece from many alpacas and you want to determine how closely related they are to each other in the bale. It has some relevance if samples were taken from multiple sites on an alpaca called a grid sample. The reason CV can be misleading is simple and best understood by example. If I tell you CV is 20 but nothing more this may sound good but lets see what happens using AFD and SD. If we have AFD 20 an SD 4 or AFD 30 and SD 6 both these provide a CV of 20. Two very different fleeces and very different end uses for each fleece. One last point to make regarding AFD, SD, and CV. A single fleece can vary. If you pull the sample from the 3rd rib and down about 5 inches (roughly midside) you may get a much different reading if your sample were taken 12 inches down rather than 5 inches. This variation can be as much as we saw in the discussion of CV which is one reason a grid sample is more revealing than a single pull sample.