ST. LOUIS -- Even 200 years ago, America's most famous explorers knew a magnetic compass doesn't necessarily point north.

But in their journey toward the Pacific Ocean, Capt. Meriwether Lewis and William Clark didn't figure out what impact the difference would have on the maps they created.

Two centuries later, Washington University geologist Bob Criss is using the explorers' journals to provide historians a way to more accurately plot Lewis and Clark's exact route.

Earth's magnetic field tends to drift over time. Only rarely do the constantly shifting magnetic poles actually line up with the true north or south poles.

In 1803, President Thomas Jefferson told Lewis and Clark to record the "variations of the compass" as they mapped their westward journey.

Criss took the explorers' 2,000-page account and found places where the pair measured solar position or stars for basic navigation and subjected the readings to rigorous calculations.

The calculations Lewis and Clark could have done "are not impossible -- basically, trigonometry -- but they just didn't have time," Criss told the St. Louis Post-Dispatch.

Geologist Duane Champion at the U.S. Geological Survey said Criss' work adds valuable data to a limited record.

The oldest record of declination in the interior United States was a map compiled in 1850 by government scientists.

"What Bob has done is gone back and add an extra 50 years," Champion said.

Criss, whose work was published in this month's issue of GSA Today, said he was impressed with how much care Lewis and Clark took in their navigational readings. The explorers' figures are accurate to a few hundredths of a degree or to an exact second, despite arduous conditions.

"It's very impressive to me that these guys, despite a treacherous journey, were able to record measurement to an accuracy that is still useful today," Criss said.