spencer hsu broke the story in today's washington post. he joins us now from washington. so i say this is a watershed moment in one of the country's largest forensic >> what has been found has been as you say that for more than two decades nearly every examiner and nearly every criminal trial in which fbi experts gave testimony against criminal defendants they overstated the strength the the significance of a match. >> srennivasan: so you say said that about a quarter of all the wrongful con vixes of people that have been exonerated later on the testimony of hair examiners or bite mark comparisons have actually helped sway juries and judges. >> that's right. out of about 329 dna examinations more than a quarter involved invalid forensic science. one of the issues here is that unlike dna whiches with developed by scientists for scientists, a lot of the earlier pattern-based techniques comparing hair fiber bite marks even tracing bullets to being fired from specific weapons were developed in the lab by law enforcement. >> srennivasan: let's just say, for example if a defens