Abstract
Americans spend >100 billion dollars on restaurant fast food each year; fast food meals comprise a disproportionate amount of both meat and calories within the U.S. diet. We used carbon and nitrogen stable isotopes to infer the source of feed to meat animals, the source of fat within fries, and the extent of fertilization and confinement inherent to production. We sampled food from McDonald's, Burger King, and Wendy's chains, purchasing >480 servings of hamburgers, chicken sandwiches and fries within geographically distributed U.S. cities: Los Angeles, San Francisco, Denver, Detroit, Boston, and Baltimore. From the entire sample set of beef and chicken, only 12 servings of beef had δ13C < −21‰; for these animals only was a food source other than corn possible. We observed remarkably invariant values of δ15N in both beef and chicken, reflecting uniform confinement and exposure to heavily fertilized feed for all animals. The δ13C value of fries differed significantly among restaurants indicating that the chains used different protocols for deep-frying: Wendy's clearly used only corn oil, whereas McDonald's and Burger King favored other vegetable oils; this differed from ingredient reports. Our results highlighted the overwhelming importance of corn agriculture within virtually every aspect of fast food manufacture.
Keywords: diet, stable isotope
By purchasing and eating 1 serving of the substrates of this study (i.e., 1 hamburger, 1 chicken sandwich, and 1 small order of fries), the consumer has gained 50% of that day's recommended calories, 80% of carbohydrates, 75% of protein (90% if the consumer is a woman), and the full day's limit of dietary fat at a cost of $3* [based on the National Academy of Sciences 2005 Dietary Reference Intakes Series (www.nap.edu) and the McDonald's Dollar Menu (www.mcdonalds.com/usa/eat/features/dollar.html)]. As meat consumption has skyrocketed in the United States, the consumption of fast food has increased disproportionately*. The production of fast food meat is a unique problem in cost-optimization: to accelerate tissue production in animals, calorie consumption is maximized, and calorie expenditure is minimized. We turned to carbon and nitrogen stable isotopes to tell us the origin of the animals' diet, based on classical models (1, 2) and observations of the conspicuous 13C signature of corn [Zea mays (3)]. Multiple studies have used δ13C value to infer the origin of meat based on assumptions of corn-based diet in North American (4–6) and conventional retail [vs. “organic” (7, 8)] farmed animals. We sought to answer the following question: What can the δ13C and δ15N of fast food beef, chicken, and fries tell us about the process of food production and the ingredients within the products?
Results
Fastfood was purchased from America's top 3 chains: McDonald's, Burger King, and Wendy's; each restaurant was sampled at 3 locations within 6 major U.S. cities: Los Angeles, San Francisco, Denver, Detroit, Boston, and Baltimore [supporting information (SI) Table S1]. At each location, 9 items were purchased: 3 hamburgers, 3 chicken sandwiches, and 3 orders of fries. Fig. 1 depicts the results of all stable isotope analyses according to the location of food purchase within the U.S.; within restaurant, servings of all foods were isotopically homogeneous (summarized in Table 1). Carbon stable isotope values in beef showed the largest total variability (11.8‰), compared with ranges in δ13C value in both chicken (2.7‰) and fries (4.9‰). Burger King beef exhibited the largest variability (10.7‰), whereas the variability within McDonald's beef (7.1‰) and Wendy's beef (6.3‰) was considerably lower. Although the maximum δ13C value found for beef was similar (within 1‰) for all chains, the minimum value for Burger King beef (−25.5‰) was ≈4‰ lighter than the minimum values for the other chains. Statistical analyses showed that carbon isotope signature of beef was significantly different between chains (Table 2): After adjusting for geographic region (Model 3), McDonald's and Wendy's beef were of significantly higher δ13C value, by 3.5 and 5.2‰, respectively. After similar adjustment for chain, West coast beef was found to be significantly depleted relative to East coast beef by 1.5‰. As for carbon isotopes in chicken, maximum and minimum values compared across all chains did not differ by >1‰. McDonald's chicken was slightly, but significantly, depleted compared with Burger King chicken (by 0.3‰), after adjustment for adjustment for geographic region. Otherwise, no significant difference was seen in the δ13C value of chicken from different chains, or different geographical regions.
Fig. 1.
Carbon and nitrogen stable isotope values of all foods sampled, positioned according to geographic region (A–E). Chains are designated by symbol, stable isotope value is designated by color. The 3 restaurants sampled differ by horizontal position, whereas the 3 servings differ by vertical position, creating a grid of 9 values for each chain within each city. Three missing points (C–E) reflect 2 lost substrates from within the entire study (1 serving of chicken and 1 serving of fries).
Table 1.
Summary of δ13C and δ15N values of fastfood sampled
Mean, ‰ | Range, ‰ | Minimum, ‰ | Maximum, ‰ | |
---|---|---|---|---|
All samples | ||||
Beef (n = 162) | ||||
δ13C | −18.0 | 11.8 | −25.5 | −13.7 |
δ15N | 6.1 | 2.1 | 5.3 | 7.4 |
Chicken (n = 161) | ||||
δ13C | −17.5 | 2.7 | −19.3 | −16.6 |
δ15N | 2.3 | 1.3 | 1.5 | 2.8 |
Fries (n = 161) | ||||
δ13C | −26.9 | 4.9 | −28.8 | −23.9 |
δ15N | ||||
McDonald's | ||||
Beef (n = 54) | ||||
δ13C | −17.7 | 7.1 | −21.7 | −14.6 |
δ15N | 6.0 | 1.2 | 5.3 | 6.5 |
Chicken (n = 54) | ||||
δ13C | −17.7 | 2.2 | −19.1 | −16.9 |
δ15N | 2.2 | 1.3 | 1.5 | 2.8 |
Fries (n = 54) | ||||
δ13C | −27.2 | 2.5 | −28.5 | −26.1 |
δ15N | ||||
Burger King | ||||
Beef (n = 55) | ||||
δ13C | −20.8 | 10.7 | −25.5 | −14.9 |
δ15N | 6.4 | 1.8 | 5.6 | 7.4 |
Chicken (n = 53) | ||||
δ13C | −17.4 | 2.7 | −19.3 | −16.6 |
δ15N | 2.5 | 0.9 | 1.9 | 2.8 |
Fries (n = 54) | ||||
δ13C | −28.0 | 1.8 | −28.8 | −27.0 |
δ15N | ||||
Wendy's | ||||
Beef (n = 53) | ||||
δ13C | −15.5 | 6.3 | −20.0 | −13.7 |
δ15N | 6.0 | 1.0 | 5.6 | 6.6 |
Chicken (n = 54) | ||||
δ13C | −17.3 | 1.9 | −18.7 | −16.8 |
δ15N | 2.3 | 0.7 | 1.9 | 2.6 |
Fries (n = 53) | ||||
δ13C | −25.5 | 3.8 | −27.7 | −23.9 |
δ15N |
For a complete list of isotope measurements, see Table S2.
Table 2.
Statistical associations within δ13C and δ15N value of foods sampled
Model 1 |
Model 2 |
Model 3 |
||||
---|---|---|---|---|---|---|
Δ, ‰ | P | Δ, ‰ | P | Δ, ‰ | P | |
δ13C | ||||||
Beef | ||||||
Chain | ||||||
Burger King (reference) | ||||||
Wendy's | 6.0 | <0.001* | 5.2 | <0.001* | ||
McDonald's | 3.4 | <0.001* | 3.5 | <0.001* | ||
Region | ||||||
East Coast (reference) | ||||||
Midwest | 2.3 | <0.001* | 0.1 | 0.657 | ||
West Coast | −0.9 | <0.001* | −1.5 | <0.001* | ||
Chicken | ||||||
Chain | ||||||
Burger King (reference) | ||||||
Wendy's | 0.5 | 0.604 | 0.5 | 0.604 | ||
McDonald's | −0.3 | <0.001* | −0.3 | <0.001* | ||
Region | ||||||
East Coast (reference) | ||||||
Midwest | 0.2 | 0.171 | 0.2 | 0.178 | ||
West Coast | 0.2 | 0.133 | 0.2 | 0.120 | ||
Fries | ||||||
Chain | ||||||
Burger King (reference) | ||||||
Wendy's | 1.5 | <0.001* | 2.4 | <0.001* | ||
McDonald's | 1.0 | <0.001* | 1.9 | <0.001* | ||
Region | ||||||
East Coast (reference) | ||||||
Midwest | 0.7 | <0.001* | 0.4 | <0.001* | ||
West Coast | 0.8 | <0.001* | 0.4 | <0.001* | ||
δ15N | ||||||
Beef | ||||||
Chain | ||||||
Burger King (reference) | ||||||
Wendy's | −0.3 | <0.001* | −0.4 | <0.001* | ||
McDonald's | −0.3 | <0.001* | −0.5 | <0.001* | ||
Region | ||||||
East Coast (reference) | ||||||
Midwest | 0.3 | <0.001* | −0.9 | <0.001* | ||
West Coast | −0.2 | <0.001* | −0.4 | <0.001* | ||
Chicken | ||||||
Chain | ||||||
Burger King (reference) | ||||||
Wendy's | −0.2 | 0.009* | −0.2 | 0.009* | ||
McDonald's | −0.3 | <0.001* | −0.3 | <0.001* | ||
Region | ||||||
East Coast (reference) | ||||||
Midwest | 0.1 | 0.484 | 0.1 | 0.426 | ||
West Coast | 0.1 | 0.281 | 0.1 | 0.309 |
*, P < 0.01.
For nitrogen isotopes in meat products, variability was constrained, relative to carbon isotopes. The entire range in δ15N value of all beef sampled was <2.2‰; for δ15N value of chicken, the range was even smaller (<1.4‰). After adjustment for geographic region, both Wendy's and McDonald's beef were found to be significantly depleted in 15N compared with Burger King beef, but by values comparable to the small amount of variability in the substrates (0.4 and 0.5‰, respectively). Similarly, after adjustment for chain, Midwest and West Coast beef was found to be significantly depleted in 15N compared with East Coast beef (0.9 and 0.4‰, respectively). For chicken, no significant difference in δ15N value was associated with geographical region; slight, but significant, differences were found in Wendy's and McDonald's chicken relative to chicken from Burger King (−0.2 and −0.3‰, respectively), after adjustment for geographic region.
French fries are composed of starch fried in lipid, and contain negligible protein, therefore only carbon isotope analysis could be performed on fries. All fries sampled ranged in 13C composition across 4.9‰, within significant and relatively large differences between chains. Both Wendy's and McDonald's fries were found to be enriched in 13C compared with Burger King fries, even after adjustment for geographical region (by 2.4 and 1.9‰, respectively). Lesser differences can be significantly attributed to geographical region: After adjustment for chain, Midwest and West Coast fries were found to be enriched in 13C compared with East Coast fries (both by 0.4‰).
Discussion
Fastfood corporations do not raise livestock, but instead buy it from other companies. Birth, growth, and slaughter are distinct events occurring at different facilities, often under different companies. Each fast food chain employs distributor companies: These suppliers organize and broker the production and transport of meat to the site of food fabrication and sale. In this way, distributors act as a barrier to consumer information†; suppliers relevant to this study provide little information beyond their use of “local farms” that feed “mixed grains.” The distributor for McDonald's is Martin–Brower, L.L.C.; Burger King and Wendy's employ the same distributor, Maines Paper and Food Service, Inc. These differences probably drive the significant differences in 13C content of beef among chains, and between West Coast and other restaurants (Table 2). In contrast, all chicken is distributed to each chain by the same company, Tyson Foods, Inc.; the extreme homogeneity seen in chicken 13C content across all aspects of the study (Tables 1 and 2; Fig. 1), speaks to the virtually identical process of chicken production for the majority of American fast food.
Most of the tissue of meat animals is constructed during the final weeks before slaughter: This is also the period when stable isotope value is set within muscle and lipid tissue. For example, δ13C in muscle and lipid tissue in cows fed on corn silage for the final 6 months of life were 7.0 and 8.1‰ heavier (respectively) than those fed on grass silage for the same period (9). In addition, δ13C in animal tissues has been positively correlated with percentage of corn in diet (7, 10–13). For beef cattle, a final diet of corn silage yielded δ13C = −18.1 and −21.1‰ for lipid and muscle, respectively (9). In contrast, a final diet of grass silage yielded average δ13C = −29.8 and −24.9‰ for lipid and muscle, respectively, and intermediate diets yielded intermediate values (9, 10). For chicken, corn-meal diet yielded breast meat with δ13C = −16.6‰, whereas “cereal”-meal diet yielded breast meat with δ13C = −26.8‰ (5). Based on a comparison with these values, 100% of the chicken and 93% of the beef sampled in this study had δ13C value consistent with an exclusively corn-based diet. From the entire study, only 12 servings of beef had δ13C < −21‰; for these animals only was an additional food source besides corn possible. We note that all 12 servings were purchased at Burger King restaurant at West Coast locations.
Besides the pioneering studies of Schmidt et al. (4) and Bahar et al. (8), empirical δ15N datasets have not been reported for meat. Beef from Europe, Brazil, and the United States ranged in δ15N value from 4.8 to 9.8‰ (4); beef raised in Ireland ranged in δ15N value from 6.2 to 7.2‰ (8); both studies suggest that high δ15N values reflect “system-wide” enrichment of 15N because of addition of mineral fertilizers. Schwertl et al. (7) found the δ15N value of cattle hair positively correlated with stocking rate [kilograms of live-weight per hectare], and also showed that stocking rate is well-correlated with N-input surplus (r2 = 0.78). Fertilization required for corn production renders corn seed and silage enriched in 15N by 2 to 3‰ (on average), compared with both natural vegetation (14) and other feeds (Table 3). Beef produced in confinement was enriched by up to 0.8‰ in 15N compared with animals raised outdoors (8). The high δ15N values of fast food beef determined by this study (average δ15N = 6.1‰; Table 1), resulted from the confinement and fertilized feed necessary for rapid tissue production. Bahar et al. (8) also commented on the “remarkably invariant” values of δ15N in confined animal meat, which we have observed for both beef and chicken (Tables 1 and 2). We speculate that chicken exhibited lower δ15N values than beef (average δ15N = 2.3‰; Table 1) because chicken are not ruminants, and therefore digest feed in only 1 step. However, the relatively high and invariant (Tables 1 and 2) values of δ15N in chicken meat also reflect the monotony and confinement of tissue production.
Table 3.
Carbon and nitrogen stable isotope composition of animal feeds
Feed | δ13C, ‰ | δ15N, ‰ | Source or ref. |
---|---|---|---|
Whole plant | |||
Alfalfa (Medicago sativa) | −28.3 to −27.4 | −1.5 to 0.3 | 7, 12 |
Grass (unspecified) | −29.3 to −27.9 | −0.3 to 4.3 | 7, 22 |
Wheat (Triticum aestivum) | −26.7 to −22.9 | 2.1 to 2.5 | 7, 12, 16 |
Potato (Solanum tuberosum) | −25.8 | −1.7 to 2.6 | 16, 23 |
Total range of above | −29.3 to −22.9 | −0.3 to 4.3 | |
Seed | |||
Triticale (Triticale secale) | −27.7 | 4.2 | 7 |
Field pea (Pisum sativum) | −27.5 | 1.8 | 7 |
Broad beans (Vicia faba) | −27.6 | 1.3 | 7 |
Barley (Hordeum vulgare) | −27.8 to −22.2 | 0.8 to 6.3 | 7, 12, 16 |
Sunflower (Helianthus Annuus) | −26.2 | 4.3 | 12, 16 |
Soy (Glycine max) | −27.4 to −25.4 | 1.1 | 12, 16 |
Oats (Avena sativa) | −25.1 to −24.1 | 2.2 | 12, 16 |
Acorns (Quercus spp.) | −24.8 to −21.0 | n.d. | 12 |
Rapeseed (″canola″) (Brassica napus) | −35.6 to −32.3 | n.d. | (Brassicaceae) 24 |
Total range of above | −35.6 to −21.0 | 0.8 to 6.3 | |
Corn (Zea mays) | −13.1 to −10.7 | 3.4 to 5.7 | 12, 16, 23, 25 |
Silage and meal | |||
Alfalfa-based | −29.0 | 2.4 | 7 |
Ryegrass-based (Lolium sp.) | −27.9 | 3.9 | 7 |
Grass (unspecified) | −29.6 to-27.7 | −0.2 to 4.0 | 7, 9 |
Soybean-based | −26.0 to −25.1 | 0.1 to 1.5 | 7, 22 |
Rapeseed-based | −26.6 | 2.7 to 2.9 | 7, 22 |
Sugarbeet-based (Beta vulgaris) | −28.3 to −28.0 | 3.6 to 5.2 | 7, 22 |
Total range of above | −29.6 to −25.1 | −0.2 to 5.2 | |
Corn-based | −14.0 to −11.0 | 4.5 to 7.2 | 7, 9, 13 |
n.d., not determined.
The result that the δ13C value of fries differed significantly among restaurants (Table 2) led us to test the hypotheses that different chains employ different protocols for deep-frying. French fries are manufactured by immersing preformed emulsified potato (Solanum tuberosum) meal in ≈190 °C fat for ≈3 min. The resulting product is mostly composed of potato, fat (17.6, 15.5 and 22.9 mass percentage for Burger King, McDonalds, and Wendy's, respectively), and minor additives [some with major consequences (15)], with negligible amounts of protein and sugar. Elsewhere we have documented the δ13C value of commercially available potatoes [δpotato = −25.8‰ (16)]; based on this value we can estimate the carbon isotope composition of the fat added to potatoes (δfat) based on the fraction of mass as fat (ffat) reported by each chain:
![]() |
This calculation assumes that potato and fat are the major mass and isotopic contributors to fries, an assumption confirmed by the negligible amount of nitrogen we found in fries (< 1% by mass) upon combustion, and by the ingredient reports available. The results of this calculation are show in Fig. 2; δ13C ranges for corn oil (−16.4 to −13.7‰) and other vegetable oils (−32.4 to −25.4‰), with cottonseed, rapeseed, soybean, sunflower, sesame, peanut, olive and palm oil are also specified (17). The statistically significant difference in the δ13C value of fries between chains implied a corn-oil based protocol for Wendy's, whereas McDonald's and Burger King favored other vegetable oils. Although Burger King reports soybean oil as the only oil used in their fries, both Wendy's and McDonald's report that their fries “may contain one of more” of canola, soy, cottonseed or corn oil.
Fig. 2.
Calculated value of δ13C in fat plotted against measured value of δ13C in fries; carbon stable isotope ranges reported for corn oil and other vegetable oils shaded for comparison (17). We note that variable %-fat within servings may account for the wide spread in δ13C of fat calculated for each restaurant.
Conclusions
Fast food corporations, although they constitute more than half the restaurants in the U.S. and sell more than 1 hundred billion dollars of food each year (18), oppose regulation of ingredient reporting‡. Ingredients matter for many reasons: U.S. corn agriculture has been criticized as environmentally unsustainable (19) and conspicuously subsidized (20). Of 160 food products we purchased at Wendy's throughout the United States, not 1 item could be traced back to a noncorn source. Our work also identified corn feed as the overwhelming source of food for tissue growth, hence for beef and chicken meat, at fast food restaurants. We note that this study did not include an examination of beverages served, which are dominantly sweetened with high fructose corn syrup (21). In 2002, the European Union adopted Regulation 178 (11) requiring suppliers to trace the origin of materials used for production. At this time in the United States, such tracing is voluntary and seldom-invoked. [A description of the U.S. Department of Agriculture National Animal Identification System is at http://animalid.aphis.usda.gov/nais/.] Our work highlights the absence of adequate consumer information necessary to facilitate an ongoing evaluation of the American diet.
Materials and Methods
For each serving of meat, charred material or remnants of bun were removed by scraping away ≈2 mm of the outer surfaces of the patty/fillet at the subsampling locations before collection. Each patty/fillet was then subsampled in 4 places, then combined to produce an averaged sample. For each serving of fries, 3 individual fries were combined to create an averaged sample. All samples were freeze-dried and homogenized by hand, using a mortar and pestle. Samples were analyzed for carbon and nitrogen composition, using a Eurovector Elemental Analyzer configured with a Micromass Stable Isotope Ratio Mass Spectrometer; values are reported in standard δ-notation relative to VPDB and AIR (Table S2). Measurement of CO2 and N2 upon combustion revealed homogeneous carbon and nitrogen content of the substrates: burger patties = 60 ± 2%C and 9 ± 1%N; chicken fillets = 46 ± 2%C and 13 ± 1%N; fries = 49 ± 2%C and <1%N (μ ±σ; the low N-content of fries precluded δ15N analysis). A summary of isotopic results is presented in Table 1.
Because samples are considered clustered within restaurants and cities, we adopted a 3-level statistical model to evaluate associations within δ13C and δ15N value of foods sampled. Each model is a simple/multivariate linear regression: The analysis was performed separately for δ13C and δ15N in beef, chicken, and fries. Model 1 evaluated the association of isotope composition with fast-food chain; model 2 evaluated the association of isotope composition with geographical region; model 3 evaluated the association of isotope composition with fast-food chain and geographical region together. Estimates (Δ) of coefficients for covariates were calculated from the following linear models (Table 2):
Model 1.
Δ = Iref + (β1 × δWendy's) + (β2 × δMcDonald's) + RIrestaurant + RIcity.
Model 2.
Δ = Iref + (β1 × δMidwest) + (β2 × δWest Coast) + RIrestaurant + RIcity.
Model 3.
Δ = Iref + (β1 × δWendy's) + (β2 × δMcDonald's) + (β3*δ Midwest) + (β4 × δWest Coast) + RIrestaurant + RIcity.
Models 1, 2, and 3 were referenced against Burger King, East Coast, and both Burger King and East Coast values, respectively. Intercepts represented 〈δ〉 for the reference group (I), or random values based on normal distribution (RI); slopes (β) were calculated separately for each model; δ represented the δ13C or δ15N of the food substrate.
Supplementary Material
Acknowledgments.
We thank the participants who donated their time, travel, and expertise; J. M. Cotton, K. J. Foebler, and W. M. Hagopian for purchasing fast food samples and for laboratory help; the Department of Earth and Planetary Sciences at The Johns Hopkins University, where the isotopic analyses took place; and E. Yeung and H. C. Hsu of the Johns Hopkins School of Public Health for statistical analyses and interpretation.
Footnotes
The authors declare no conflict of interest.
This article contains supporting information online at www.pnas.org/cgi/content/full/0809870105/DCSupplemental.
Total meat consumption has increased 63% between 1950 and 2005 in the United States according to the USDA 2008 United States per capita food availability report (www.ers.usda.gov/data/FoodConsumption/FoodAvailQueriable.aspx); total consumption by food type can be found within the 2007 USDA food consumption and nutrient intake tables (www.ers.usda.gov/data/FoodConsumption/FoodAvailSpreadsheets.htm).
Bahar et al. (ref. 8, p 1300) provides an example of this: “This survey was originally designed to relate patterns in isotopic compositions of beef to geographical and husbandry background information; however, because of a lack of disclosure of farm-level information by the suppliers, the interpretation is mainly restricted to the description of overall seasonality of Irish organic and conventional beef.”
“Providing calories, or any other nutrition information, should be an education issue, not a political one” (www.wendys.com/food/pdf/us/menuboard.pdf).
References
- 1.DeNiro MJ, Epstein S. Influence of diet on the distribution of carbon isotopes in animals. Geochim Cosmochim Acta. 1978;42(5):495–506. [Google Scholar]
- 2.DeNiro MJ, Epstein S. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim Cosmochim Acta. 1981;45:341–351. [Google Scholar]
- 3.Farquhar GD. On the nature of carbon isotope discrimination in C4 species. Australian J Plant Physiol. 1983;10:205–226. [Google Scholar]
- 4.Schmidt O, et al. Inferring the origin and dietary history of beef from C, N and S stable isotope ratio analysis. Food Chem. 2005;91:545–549. [Google Scholar]
- 5.Morrison DJ, Dodson B, Slater C, Preston T. 13C natural abundance in the British diet: Implications for 13C breath tests. Rapid Commun Mass Spectrom. 2000;14:1321–1324. doi: 10.1002/1097-0231(20000815)14:15<1321::AID-RCM946>3.0.CO;2-8. [DOI] [PubMed] [Google Scholar]
- 6.Schoeller DA, Klein PD, Watkins JB, MacLean WC. 13C abundances of nutrients and the effect of variations in 13C isotopic abundances of test meals formulated for 13CO2 breath tests. The Am J Clin Nutr. 1980;33:2375–2385. doi: 10.1093/ajcn/33.11.2375. [DOI] [PubMed] [Google Scholar]
- 7.Schwertl M, Auerswald K, Schäufele R, Schnyder H. Carbon and nitrogen stable isotope composition of cattle hair: Ecological fingerprints of production systems? Agric Ecosyst Environ. 2005;109:153–165. [Google Scholar]
- 8.Bahar B, et al. Seasonal variation in the C, N and S stable isotope composition of retail organic and conventional Irish beef. Food Chem. 2008;106:1299–1305. [Google Scholar]
- 9.Bahar B, et al. Alteration of the carbon and nitrogen stable isotope composition of beef by substitution of grass silage with maize silage. Rapid Commun Mass Spectrom. 2005;19:1937–1942. doi: 10.1002/rcm.2007. [DOI] [PubMed] [Google Scholar]
- 10.De Smet S, Balcaen A, Claeys E, Boechx P, Van Cleemput O. Stable carbon isotope analysis of different tissues of beef animals in relation to their diet. Rapid Commun Mass Spectrom. 2004;18:1227–1232. doi: 10.1002/rcm.1471. [DOI] [PubMed] [Google Scholar]
- 11.Boner M, Förstel H. Stable isotope variation as a tool to trace the authenticity of beef. Anal Bioanal Chem. 2004;378:301–310. doi: 10.1007/s00216-003-2347-6. [DOI] [PubMed] [Google Scholar]
- 12.González-Martin I, González-Pérez C, Méndez JH, Marques-Macias E, Sanz-Poveda F. Use of isotope analysis to characterize meat from Iberian-breed swine. Meat Science. 1999;52:437–441. doi: 10.1016/s0309-1740(99)00027-3. [DOI] [PubMed] [Google Scholar]
- 13.Piasentier E, Valusso R, Camin F, Versini G. Stable isotope ratio analysis for authentication of lamb meat. Meat Science. 2003;64:239–247. doi: 10.1016/S0309-1740(02)00183-3. [DOI] [PubMed] [Google Scholar]
- 14.Amundson R, et al. Global patterns of the isotopic composition of soil and plant nitrogen. Global Biogeochem Cycles. 2003;17:1031. [Google Scholar]
- 15.Associated Press. French Fry Fracas: McDonald's Apologizes For “Confusion” Over Beef Flavor In Fries. 2001 May 25; [television news broadcast] (CBS News) [Google Scholar]
- 16.Jahren AH, et al. An isotopic method for quantifying sweeteners derived from corn and sugar cane. Am J Clin Nutr. 2006;84(5):1380–1384. doi: 10.1093/ajcn/84.6.1380. [DOI] [PubMed] [Google Scholar]
- 17.Kelly SD, Rhodes C. Emerging techniques in vegetable oil analysis using stable isotope ratio mass spectrometry. Grasas y Aceites. 2002;53(1):34–44. [Google Scholar]
- 18.United States Census Bureau. Washington, D.C: U.S. Census Bureau; 2002. Economic Census. Available at www.census.gov/econ/census02/ [Google Scholar]
- 19.Tilman D. The greening of the green revolution. Nature. 1998;396:211–212. [Google Scholar]
- 20.Horrigan L, Lawrence RS, Walker P. How sustainable agriculture can address the environmental and human health harms of industrial agriculture. Environmental Health Perspectives. 2002;110:445–456. doi: 10.1289/ehp.02110445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Popkin BM, et al. A new proposed guidance system for beverage consumption in the United States. Am J Clin Nutr. 2006;83(3):529–542. doi: 10.1093/ajcn.83.3.529. [DOI] [PubMed] [Google Scholar]
- 22.Fuller BT, et al. Nitrogen balance and δ15N: Why you're not what you eat during nutritional stress. Rapid Commun Mass Spectrom. 2005;19:2497–2506. doi: 10.1002/rcm.2090. [DOI] [PubMed] [Google Scholar]
- 23.Petzke KJ, Boeing H, Klaus S, Metgas CC. Carbon and nitrogen stable isotopic composition of hair protien and amino acids can be used as biomarkers for animal-derived dietary protein intake in humans. The J Nutr. 2005;135:1515–1520. doi: 10.1093/jn/135.6.1515. [DOI] [PubMed] [Google Scholar]
- 24.Jahren AH, Arens NC, Harbeson SA. Prediction of atmospheric δ13CO2 using fossil plant tissues. Reviews of Geophysics. 2008;46(RG1002) doi: 10.1029/2006RG000219. [Google Scholar]
- 25.Bol R, Pflieger C. Stable isotope (13C, 15N, and 34S) analysis of the hair of modern humans and their domestic animals. Rapid Commun Mass Spectrom. 2002;16:2195–2200. doi: 10.1002/rcm.706. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.