The most widely accepted theories about the development of our solar system are based on a particular scheme -- An ancient disk of gas and dust formed several small planetary embryos. They collided and coalesced over time, first gathering together their rocky cores and then, for some, a gaseous or icy shell.

But when scientists run various versions of this model on their computers, the solar system they create is missing a couple of important members -- Uranus and Neptune. That can be frustrating for a grown-up researcher, coming up with only seven planets when children are taught there are nine.

Existing models suggest that the transition from a protoplanetary nebula (the ancient disk of gas and dust) to a collection of planets was somewhat orderly -- though along the way there would have been many violent crashes between small bodies, and also some larger collisions. The thinking is that planets collected their material gradually while carving arcs that pretty much match their present-day orbits.

The problem with that idea, however, is that there wasn't much out where Uranus and Neptune are located at 1.78 billion and 2.79 billion miles (286 billion and 449 billion kilometers) from the sun, respectively. The protoplanetary disk is thought to have become thin and meager as distance from the nascent sun increased.

So existing theories hold that these two large planets, suspected to have rocky cores but otherwise composed mostly of ice, could not have gathered enough ingredients to become so massive in the time frame during which the other planets are believed to have formed.

In other words, Uranus and Neptune couldn't have formed as quickly as they did, unless they spent their formative period elsewhere.

Martin Duncan and his colleagues propose this theory.

"We realized that perhaps Uranus and Neptune actually formed closer to the sun, in the same region as Jupiter and Saturn," said Duncan, of Queen's University in Kingston, Canada.

Duncan paints a protoplanetary picture with four somewhat huddled embryonic cores, Jupiter being the closest to the sun and perhaps already slightly larger than the others. Presently, Jupiter is 483 million miles (777 million kilometers), while Saturn is 886 million miles (1425 million kilometers), from the sun.

Being close, Jupiter had the most material to work with, Duncan said. When it reached a certain size, about five to 15 times the mass of Earth, Jupiter's gravity began quickly pulling in surrounding gas and soon became the mostly gaseous giant that it is today. In this scenario, it is likely that something similar happened to Saturn.

It was the sudden presence of another giant or two that threw the whole system into chaos, Duncan said.

Uranus and Neptune were flung outward, as though from a slingshot, by the sudden introduction of nearby mass. Their orbits became erratic, swinging into an egg-shaped pattern and slipping above and below the imaginary plane along which most planets move around the sun. Disorder reigned for hundreds of thousands of years -- as Uranus and Neptune gathered up icy comets in the far reaches of the developing solar system. Finally, the two ice giants settled into their present configurations.

There are other ideas about how Jupiter formed. One is that it migrated in from farther out in the solar system. Both ideas can't be accurate, Duncan admits, but his is the first that explains how Uranus and Neptune came to be.

Higher-powered computer modeling will be necessary to determine which theory is correct. Like any computer model, there are limitations to the one Duncan and his colleagues ran. He said some possible effects of interplanetary gas were left out of the equations, and those variables could alter how the early planets interacted.

"The solution is only partial because one still needs to explain how the Jupiter and Saturn cores accreted large amounts of gas whereas Uranus and Neptune did not," writes Run Malhotra in an accompanying analysis in Nature. Malhotra, of the Lunar and Planetary Institute in Houston, also questions the highly dynamic nature of the new theory, suggesting that a less-chaotic method of planetary formation is still conceivable.

"We're just learning about giant planet formation," Duncan says. "There's a lot more to be understood about it."