Refering to today's lube-tips question regarding identifying higher particle counts, I have experienced the same thing. I have not done additional ferrous density or analytical ferrography testing when this has occurred. Based on my experience, I can't justify the added expense. In the past, I have filtered the component oil through a kidney-loop system and achieved my target cleanliness - However, I have also simply re-sampled. Each time, the results have been surprising cleaner in the second sample. I question sampling technique first, but I also question things that I cannot control, such as sample bottle cleanliness, and oil lab techniques. I also wonder about microscopic entrained air bubbles.