The idea is emotionally powerful: modern agriculture has stripped the soil, vegetables have lost their minerals, and we would need to eat several oranges to obtain what one orange provided fifty years ago. Parts of that story are supported by evidence. The neat, universal version is not.

Food composition is not fixed. Variety, soil, climate, carbon dioxide, fertilisation, harvest time, storage, ripeness, water content and laboratory method can all change the result. A tomato is not one chemical object across every farm and decade. That makes historical comparison difficult, but not meaningless.

What the historical data found

A widely cited analysis compared United States food composition data for 43 garden crops between 1950 and 1999. Across the group, statistically reliable declines appeared in protein, calcium, phosphorus, iron, riboflavin and vitamin C. Median declines ranged from about 6 percent for protein to 38 percent for riboflavin. Seven other nutrients did not show reliable group changes, and some individual values increased.

That is evidence of change, not proof that every modern vegetable contains less of every nutrient. Historical food tables used different sampling and analytical methods. Moisture can change the concentration measured per 100 grams. Cultivars also changed. The authors themselves treated modern breeding and the trade-off between yield and nutrient concentration as a leading explanation.

Wheat gives us a cleaner historical window

The Broadbalk experiment in England is useful because archived wheat grain and soil stretch back to the nineteenth century. Researchers found that zinc, iron, copper and magnesium in grain were relatively stable from 1845 until the mid-1960s, then declined. The timing matched the arrival of high-yield semi-dwarf cultivars. Soil mineral levels had not fallen in parallel and in some cases had risen.

This points to yield dilution. A plant can produce more carbohydrate and more grain without increasing mineral uptake at the same rate. The harvest becomes larger while the concentration of some nutrients per gram falls. That is a breeding and physiology story as much as a soil-depletion story.

Carbon dioxide changes the chemistry too

Higher atmospheric carbon dioxide can increase carbohydrate production in many crops while reducing the relative concentration of protein and minerals. Large analyses of C3 crops, including wheat and rice, have reported lower zinc and iron under elevated carbon dioxide. The effect is not identical in every crop, nutrient or growing condition, but it matters because the atmosphere is part of the food system.

What this does not mean

It does not mean vegetables and whole grains are nutritionally pointless. They remain major sources of fibre, potassium, folate, vitamin C, polyphenols and many other compounds. It does not mean a person can correct a poor diet with a handful of pills. It also does not prove that soil exhaustion is the single explanation for every measured decline.

The larger clinical problem is often the combination of modest nutrient-density changes with a much bigger dietary change: fewer legumes, vegetables, nuts, seeds and minimally processed foods, and more refined food that was never nutrient-dense in the first place. A slightly less mineral-dense carrot inside a varied whole-food diet is one issue. Replacing the carrot with ultra-processed food is another.

A practical response

Eat diversity rather than betting everything on one crop. Rotate vegetables and whole grains. Use legumes, nuts, seeds, herbs and mineral-rich foods. Pair plant iron sources with vitamin C. Choose produce with good flavour and freshness, because cultivar, maturity and storage matter. Support farming systems that protect soil structure and biological activity, but do not confuse a production label with a guaranteed nutrient analysis.

Where risk is real, test the person rather than diagnosing the entire food supply. Restrictive diets, gastrointestinal disease, pregnancy, heavy menstrual loss, medication use, ageing and low food intake can all justify individual assessment. Supplementation should answer a defined need, not a frightening graph shared without context.

Clinical takeaway

Some modern crops do contain lower concentrations of particular nutrients than older comparisons suggest. The pattern depends on crop, cultivar, nutrient and environment. Build a varied, minimally processed diet first, assess individual risk, and use targeted supplementation when there is a reason.

References used for fact-checking