Kenneth Rendell pulls out a folder of musty letters and notes. Inside is a trove of timeworn, historic snippets - a note from George Washington, a payment stub from Ben Franklin, a water-spotted Charles Dickens letter. Only trouble is: They're all fake.
''I have a whole collection of forgeries,'' says the document sleuth, who wears tennis shoes and works out of a Boston-area office. ''In most all cases we can authenticate a historical document.''
Mr. Rendell is one of a small fraternity practicing the arcane science of detecting forgeries and authenticating old documents.
The practice - part science, part art - has been spotlighted anew since the surfacing of diaries purported to be penned by Adolf Hitler. The black-bound, 60 -volume archive represents either a major historical find or one of the century's most ambitious hoaxes (a growing number of historians put it in the latter camp).
In a world tumbling toward razzle-dazzle technologies, document sleuthing remains a science of relatively simple tools. Clues are frequently ferreted out using magnifying glasses, microscopes, chemical analysis of ink, and ultraviolet and infrared light. But often as not, it involves old-fashioned eyeballing of the questioned documents next to authenticated ones or checking the historical references in the text.
Usually historians and document detectives need go no further than the handwriting itself. Though styles change - Hitler's writing, for instance, fluctuated widely with his moods - handwriting is a crude sort of fingerprint. By checking the loops of an ''l'' or where strokes begin and end, historians can often pick out forgeries, provided there are plenty of ''real'' samples from the same time period for comparison.
One telltale sign of fabrication is handwriting that is too good. Often it means the name has been traced or painstakingly drawn. When someone recently approached Mr. Rendell with a letter supposedly written by Zachary Taylor, he saw quickly that some letters had been written backward - retraced. Probing with a microscope, he found pencil lines under the ink where someone had copied the writing.
Pens and inks are crucial. Many an amateur forger has tried to pass off a George Washington document signed with a steel pen, when quills were the technology of the day. Other subtler variations in ink content show up under ultraviolent light or in chemical analysis. They can help identify the time period in which a note was written. Documents penned with water-soluble inks age over time. Older, acid-base inks often eventually eat through the paper. A growing number of manufacturers are putting trace elements in inks that make company and time identification easier.
Similar analysis can point up differences in the stock of paper used - and thus also help place the time period. Some forgers stain paper to try to make it look old - a trick that fools many a layman but few experts.
Often the content of the documents - references, dates, places - is the easiest way to determine authenticity.
Most forgers, not surprisingly, put pen to document for money. A photograph signed by John Kennedy might fetch $150 from collectors today, a Stalin letter $ 10,000. In other cases fame is more important than financial reward: In Victorian England, William Gladstone's signature didn't bring wealth, but it did raise impressed eyebrows at a party. Still others doctor documents for political reasons.
All three have been suggested as motives for forging the Hitler diaries. Some believe they are the work of East bloc counterfeiters trying to discredit the West or stir up political tensions within West Germany. Others theorize it could be a neo-Nazi attempt to buff up Hitler's image on the 50th anniversary of his rise to power.
On one point most experts seem to agree: If a group of handwriting specialists and historians could go over the whole collection, not just a few pages of it, the record definitely could be set straight. Certainly ample clues - ink, paper, signatures, binding glue (which can be dated), historical references - are there to make a definitive judgment.