In May 25 in Geneva, Iran backed away from its threat to resume its uranium enrichment effort, which many countries believe will lead to a nuclear weapon. In return, Britain, France and Germany agreed not to ask the United Nations to punish Iran for its past nuclear violations. While the deal undoubtedly staved off a crisis, it did nothing to answer two big questions: first, how much Iran's nuclear program has really achieved, and second, whether it is pursuing the bomb.

The answers to these questions have eluded the International Atomic Energy Agency for two years, but they are essential to resolving the Iranian nuclear crisis, no matter what the outcome of the current talks.

The Agency is still trying to puzzle out the source of several types of highly enriched uranium found in Iran. Traces of uranium enriched to approximately 70%, 54% and 36% turned up at the Kalaye Electric Company, home to Iran's secret centrifuge enrichment effort from 1995 until 2003. Particles of 54% enriched uranium were also discovered in chemical traps at the enrichment plant at Natanz, now the center of Iran's enrichment program. In addition, the IAEA uncovered particles of enriched uranium on centrifuge manufacturing equipment.

Tehran says it has only enriched uranium to a level of 1.2%. It claims that the highly enriched material arrived on parts purchased from abroad. The Agency is still trying to corroborate this story. If Iran produced the highly enriched uranium itself, it would mean that Iran is still dissembling. It would also mean that Iran had gone far beyond the enrichment level needed for civilian power reactors (around 3%) and was approaching levels suitable for a weapon (about 90%).

In addition, the Agency is investigating Iran's work on the P-2 centrifuge. The P-2 is more efficient at enriching uranium than the older P-1 centrifuge that Iran has been operating. Iran first divulged its work on the P-2 in early 2004, several months after it was supposed to provide a "correct, complete and final picture" of its nuclear work. At that time, Iran said it once had a small-scale P-2 program, but that it had been shut down. Only when pressed by the IAEA did Iran admit, several months later, to importing magnets for the P-2, and to having sought thousands more. Why such quantities for a small research program that had supposedly been mothballed? And why lie about it? The answers to these questions could reveal how much enriched uranium Iran can really produce-and how quickly.

In September 2003, the IAEA discovered that during the early 1990s Iran had attempted to produce polonium-210. This work was a red flag, as polonium is specially suited for use with beryllium to trigger nuclear weapons. Iran claimed it wanted the polonium for nuclear batteries, an explanation the IAEA found implausible. Further fueling suspicion, Iran was reported to be seeking large quantities of beryllium around the same time.

To make matters worse, the IAEA has been foiled in its attempts to track down activities at two suspect sites known as Parchin and Lavisan-Shian. The Agency wants to know whether Iran used these facilities for secret nuclear work. Parchin, a sprawling military complex near Tehran, is a likely candidate for explosive testing of nuclear components. But the Agency's inspectors were only allowed to see limited areas of Parchin in January, some eight months after the IAEA's initial request. A request for a follow-up visit has been turned down.

All hope of discovering what might have happened at Lavisan now appears lost, given the fact that Iran razed the site between November 2003 and March 2004. Iran has refused to tell the IAEA what equipment was bought for the site, saying that the equipment "could be used in conventional activities, which Iran is not obliged to declare."

The Agency is also waiting for more information on Iran's black market network of nuclear suppliers. Iran got several deliveries of centrifuge parts in the 1990s, and the IAEA wants to know exactly what Iran received, and from whom. Unraveling Iran's clandestine supplier network is as important as decoding the origin of the highly enriched uranium found in the country. It could reveal how Iran built its program and how advanced it is.

Through agile diplomacy, the talks between Iran and the Europeans may yet succeed. Tehran may agree to freeze uranium enrichment permanently. Or, as appears more likely, the present compromise will fall apart and the freeze will end. In either case, it will be essential to know how far Iran actually got in its secret nuclear work. Can Iran already make bomb-grade uranium? Can it run advanced centrifuges? Has it already worked on building a nuclear warhead? Who are its main suppliers and what have they sold? These questions will dominate the next phase of the Iranian nuclear debate, no matter what happens with the present talks. The IAEA needs to continue digging until the answers are found.