Yes, say some nutritional experts, who argue that carbohydrates should be labelled good or bad, just the way fats are, and that mashed potatoes and white bread are dietary evil.

Other equally respected nutritional experts hold that a carb is a carb is a carb.

The debate revolves around the idea of the glycemic index, which is a way of rating how quickly carbohydrates enter the bloodstream as sugar -- fast is bad -- and is offered as scientific underpinning by the authors of a variety of low-carb diet plans.

"It's almost unethical to tell people to eat a low-fat, high-carbohydrate diet with no regard to glycemic index," says Janette Brand-Miller of the University of Sydney, one of the field's pioneers.

The glycemic index is endorsed by the World Health Organization, but it remains deeply controversial.

The fact that carbohydrates break down at different rates has been known for a long time. It is why diabetics were once told to avoid sweets, since presumably sugary foods would quickly turn into sugar in the blood stream. About 20 years ago, scientists came up with the glycemic index, or GI. The body converts all carbohydrates into sugar molecules that are burned or stored. The faster carbs are broken down by the digestive system, the quicker the blood sugar goes up and the higher their GI.

The GI of at least 1,000 different foods has been measured, and the findings have been surprising. Some complex carbohydrates are digested faster than the long-demonized simple carbs. Foods such as white bread and some breakfast cereals break down in a flash, while some sweet things, like apples and pears, take their time.

In general, starchy foods like refined grain products and potatoes have a high GI -- 50 per cent higher than table sugar. Unprocessed grains, peas and beans have a moderate GI. Non-starchy vegetables and most fruits are low.

While it seems reasonable that chewy, whole-grain bread is digested more slowly than white, some of the results are less obvious. For instance, overcooking can raise the GI. Ripe fruit is lower than green. A diced potato is lower than mashed, and thick linguine is lower than thin.

To make matters even more confusing, the glycemic index measures only carbohydrates. Some vegetables, such as carrots, have quite high GIs, but they don't contain much carbohydrate, so they have little effect on blood sugar.

Therefore, some experts prefer to speak of food's glycemic load, which is its glycemic index multiplied by the amount of carbohydrates in a serving. Considered this way, a serving of carrots has a modest glycemic load of three, compared with 26 for an unadorned baked potato.

Blood sugar levels may shoot twice as high after a high-GI meal as after a low one, and that unleashes metabolic havoc: the body responds with a surge of insulin, which prompts it to quickly store the sugar in muscle and fat cells. The high sugar also inhibits another hormone, glucagon, which ordinarily tells the body to burn its stored fuel.

Blood sugar soon plunges. In fact, so much is stored so fast that within two or three hours, levels may be lower than they were before the meal. Suddenly, the body needs more fuel. But because glucagon is still in short supply, the body does not tap into its fat supply for energy. The inevitable result? Hunger.

Experiments to prove this are difficult and time-consuming. Among those trying is Dr. David Ludwig of Boston's Children's Hospital, who has done studies on overweight teenagers.

In one, he tested the idea that a high-GI breakfast makes people hungrier at lunch. A dozen obese boys were fed three different breakfasts, all with the same calories -- a low-GI vegetable omelette and fruit, medium-GI steel-cut oats or high-GI instant oatmeal.

At noon, they could eat as much as they wanted. Those who started the day with instant oatmeal wolfed down nearly twice as much as those getting the veggie omelette.

Ludwig says overweight people do not need to starve themselves. On a low-GI diet, they can eat enough to feel satisfied and still lose weight, whether they follow a standard low-fat diet or a high-fat Atkins approach.

"I think the U.S. nutritional societies are going to come to see glycemic index not as a radical new notion but actually as the optimal compromise position, one that makes perfect nutritional sense," says Ludwig.

Even such small experiments have been rare. Most support for the idea comes from big surveys that follow people's health and diets over time. Some of these show that those who consistently favour low-GI fare are less likely to become overweight or to get diabetes and heart disease.

The evidence is strong enough for authors of some popular diet books, who use the glycemic index as one of their primary rationales. "It's a new unifying concept that brings nutritional habits out of the dark ages and says it's all about the numbers," says Barry Sears, author of the Zone series of diet books. "It says diet does not have to be based on philosophy. It can be based on hard science."

The American Heart Association, an especially influential arbiter of what people should eat, disagrees. The head of its nutrition committee, Dr. Robert Eckel of the University of Colorado, says the theory that high-GI foods make people hungry is "ridiculous" and argues that a scientific case can be made for just the opposite.

"We are getting fatter because we are eating more calories," he says. "Whether that relates to glycemic index is highly debatable."

Dietitians generally encourage a balanced, varied diet emphasizing unadulterated whole foods and they cringe at a classification that puts ordinary baked potatoes and white rice on a taboo list.

"It's an artificial system of classifying foods as good and bad," says JoAnn Carson, a nutritionist at the University of Texas Southwestern Medical Centre. "There is no reason people should avoid potatoes."

Carla Wolper, a dietitian at New York Presbyterian Hospital, dismisses the idea. "Trust me," she says. "This is a scientific red herring."

Others worry that the whole business, with competing numbers for glycemic index and glycemic load, with different ratings for linguine versus rigatoni and even al dente versus well done, is just too hard to keep straight.

"We are putting before the public an extraordinarily complicated message, which I don't think they will follow or be very happy with," says Dr. Xavier Pi-Sunyer, head of obesity research at St. Luke's Roosevelt Hospital Centre in New York City.

Defenders of the idea say people can get most of the benefit just by following a few broad rules.

"I do think this is an important concept for people to understand, but I don't think they need to worry about specific numbers," says Dr. Walter Willett of the Harvard School of Public Health. "In practice, it means consuming refined starches -- white bread, white rice, white pasta -- sugary foods and beverages and potatoes sparingly."

Willett says people with diabetes should be especially careful to watch their glycemic load, which he thinks is as important as the total amount of carbohydrates they eat. His team's Nurses Health Study, following nearly 85,000 women for 16 years, found a 40-per-cent increase in the risk of diabetes in those on high glycemic load diets.

However, the diabetes association is skeptical of studies that look for trends in large population groups and concludes the evidence for the glycemic index is just too weak. "The jury is still out in regard to how glycemic index should be used in those with diabetes," says Dr. Nathaniel Clark, head of clinical affairs.

Meanwhile, some nutritional experts say that while the glycemic index may help people pick healthier foods, they caution it is only part of the formula for good eating. Yale's Dr. David Katz notes the index offers no guidance on the fat and protein people should eat.

"Our diets are too high in bad foods, refined white flour and simple sugars, but also all the wrong kinds of fat," Katz says. "It's the overall diet -- fat, protein and carbohydrates. You have to make the right choices in all of them."

HOW THE GLYCEMIC INDEX WORKS

Scientists estimate the glycemic index -- or GI -- of a food by watching what happens when people eat it.

Typically, they measure 10 volunteers' blood sugar levels after eating 50 grams of pure carbohydrate from a test food. They compare that with their blood-sugar levels after eating 50 grams of either white bread or table sugar. Responses vary, so they take the average.

The numbers may not be as precise as they seem at first. For instance, the GI of different bags of white rice can vary, because the kind of starch found in rice differs from country to country. Even the size of a potato plays a role, since starch molecules are more branched in larger potatoes, raising their GI.

Still, Janette Brand-Miller of the University of Sydney says such differences are "really splitting hairs. There is no doubt that most potatoes in any form have a high glycemic index."

The index was developed in 1981 by Dr. David Jenkins and colleagues from the University of Toronto. Because the amount of carbohydrate in food varies, researchers from Harvard University in 1997 came up with the idea of glycemic load, which compares the effect on blood sugar of typical servings.

There is no glycemic index for meat, chicken, fish, lettuce, cheese or eggs, among other things, because they contain almost no carbohydrate.

Ran with fact box "How the Glycemic Index Worksv", which has been appended to this story.