Bred or genetically modified: what is the difference?

The short answer

Breeding changes thousands of genes at once without anyone knowing which. Genetic modification changes one, in a place that is known. Untargeted against targeted.

What they deliver differs just as much. Breeding produces the varieties on your supermarket shelf: bigger, sweeter, longer keeping. Modification so far mostly produces an advantage in the field, and sits in a handful of crops you rarely eat as a vegetable.

What is breeding, exactly?

Crossing and selecting. You put two plants together, grow the offspring, and keep the one closest to what you were after. Then you do it again, sometimes for decades.

The second method is less well known and has been ordinary for eighty years: mutation breeding. Seed goes under gamma rays or into a chemical bath, the DNA breaks in random places, and you grow on whatever comes out useful. Thousands of varieties were made this way, including wheat, rice and pears that sit on the shelf unlabelled.

With both, you know what the plant does and not what happened in its DNA.

What is genetic modification, exactly?

One piece of DNA, in a place you choose. You know which gene you are adding, what it codes for and where it lands, and you can check afterwards whether it went as intended.

Around it sits an approval process that does not exist for bred varieties. A new bred variety can reach the market untested; a modified one costs years and millions in paperwork.

Which method changes the plant more?

The untargeted one. In rice they were set side by side: researchers measured gene activity in plants treated with gamma rays and in plants carrying one inserted piece of DNA, each against its own untreated original [1].

In every case studied, the change was larger in the irradiated plants. Both methods switch genes on and off that nobody intended, and the one without a label did more of it.

Their own conclusion: judge a new variety case by case, rather than on whether a gene technique was involved.

What does modification deliver?

An advantage in the field, and that is well measured. Across 147 studies of farms and trial plots, chemical pesticide use fell by 37 percent, yields rose by 22 percent, and farmer profits rose by 68 percent [2].

The gains were larger for insect-resistant crops than for herbicide-tolerant ones, and larger in developing countries than in rich ones. These are agricultural figures and not health figures. Less pesticide in the field is something that does reach your plate.

A lot of safety work has been done. Four researchers gathered a decade of literature and found no hazard directly connected with the use of these crops [3].

What does breeding deliver, and what does it cost?

It delivers everything you eat. Every carrot, tomato and apple on the shelf is a bred variety, chosen for yield, size, shelf life and transport.

What it costs has been measured. A comparison of the American food tables from 1950 and 1999 across 43 garden crops found declines in six of the thirteen nutrients examined: protein, calcium, phosphorus, iron, riboflavin and vitamin C. The median declines ran from 6 percent for protein to 38 percent for riboflavin [4].

The explanation the authors give is the varieties that replaced the old ones over that half century, where yield and nutrient content trade off against each other.

Then comes the honesty you rarely get on this subject. Crop by crop and nutrient by nutrient, most of it cannot be told apart from no change at all. Depending on the assumption, 33 or 20 percent of the ratios differed reliably from one, and about 28 percent went up instead.

What do we choose here ourselves?

Away from what humans made, unless it is proven to be better. That applies to both methods here and not to one of them.

Breeding gets no exemption for being old and carrying no label. Modification gets no conviction for being new and sounding frightening. The question is the same in each case: what was changed, why, and has anyone measured that you are better off for it.

For modified crops that gain has so far been measured in the field and not in a person. That is no reason to fear them and no reason to choose them.

What do we still not know?

Three things, and they matter.

The thinning of vegetables rests on one comparison of two tables fifty years apart, using measurement methods that changed in between. The authors say so themselves, and nobody has redone that comparison with better data since.

The safety of modification has been studied on the crops that are on the market, and those are a handful: soy, maize, cotton, rapeseed. A technique is not safe. A product is.

And what breeding has done to the taste and the contents of one tomato has been asserted far more often than it has been measured. We found no figure there we would be willing to put in print.