INVESTIGATORS OBTAINED HANDWRITING SAMPLES FROM EVERYONE INTERVIEWED, WHILE PHOTOGRAPHS OF THE QUESTIONED HANDWRITING WERE PUBLISHED IN THE PRESS AND AIRED ON TELEVISION. THE PUBLIC WAS ASKED TO SEND IN ANY HANDWRITING THAT RESEMBLED THAT ON THE ENVELOPES. ALTHOUGH A SEARCH OF SUCH MAGNITUDE COULD ONLY BE UNDERTAKEN WITH THE HELP OF UNTRAINED PERSONNEL, A PROCEDURE WAS ESTABLISHED TO ENSURE THAT AN UNTRAINED PERSON COULD NOT MAKE AN INCORRECT POSITIVE IDENTIFICATION. LABORATORY PERSONNEL SELECTED THE MOST DISTINCTIVE LETTERS AND NUMBERS TO STUDY. FOR INSTANCE, ALL THE LETTERS 'F' IN THE QUESTIONED HANDWRITING WERE WRITTEN WITH ONLY TWO STROKES OF THE PEN. THE FIRST MEMBER OF A THREE-PERSON TEAM EXAMINED THE LETTER 'F' IN THE VARIOUS SPECIMENS. IF THE LETTER WAS APPARENTLY CONSTRUCTED IN THE TWO-STROKE MANNER, THE SPECIMEN WAS PLACED INTO THE ACCEPTED GROUP. THE ACCEPTED SPECIMENS WERE EXAMINED BY THE SECOND MEMBER OF THE TEAM WHO LOOKED AT THE LETTER 'G,' ACCEPTING THOSE WITH AN S-SHAPED BODY AND NO CROSS STROKE. A THIRD MEMBER EXAMINED THE LETTER 'W' IN THE ACCEPTED SPECIMENS. AFTER THE THREE STAGES ABOUT 2 PERCENT OF THE SPECIMENS WERE ACCEPTED. LATER, EXAMINERS INTRODUCED THE CHARACTERS 'C' AND '9' AND NARROWED THE NUMBER OF SPECIMENS DOWN TO 4,900. AFTER THE ARREST, THE MURDERER PROVIDED ABOUT 3,500 OF HIS OWN WRITING SAMPLES. THIS HANDWRITING EVIDENCE CONNECTED HIM WITH THE ENVELOPES AND CONCLUSIVELY ESTABLISHED GUILT. A LARGE SCALE SEARCH OF HANDWRITING IS FEASIBLE PROVIDED SOME DEGREE OF RISK IS TOLERATED, BUT SUCH SEARCHES SHOULD NOT BE UNDERTAKEN LIGHTLY. PICTURES OF THE HANDWRITING ARE INCLUDED. (JLF)