Abstract

Weeds abound in urban and agricultural environments. Depending on region and site, up to 66% of weed species are edible, and may constitute an additional food source for humans. Based on 400 samples, ¼ m² each, collected in tropical areas (e.g., roadsides, urban vacant lots, streets, sugar cane and coffee plantations in Coatepec, Mexico), average figures of edible fresh biomass vary between 1277 and 3582 kg/ha. A similar survey performed in a temperate area (739 samples in Bariloche, Argentina) showed mean values between 287 and 2939 kg/ha. A total of 43 species were sampled in Coatepec and 32 species in Bariloche. The general means were 2.1 and 1.3 tons/ha, respectively. At a greater geographic scale, a comparison between Mexican and Argentine weeds shows that, proportionately, the food parts vary a little between regions. In general, from higher to lower, the order of uses goes from leaves, seeds, roots, fruits, herbals, flowers and condiments. Edible roots (including bulbs and rhizomes) appear to be more common among perennials than among annuals.

Key words

Argentina, Bariloche, Coatepec, edible weeds, food plants, gathering, Mexico, Patagonia, urban flora.
 

 
Although plants have sustained the hunting and gathering peoples since the Paleolithic, the prevailing knowledge about edible species began to be lost since the invention of agriculture in the Neolithic. Two well-preserved mummies found in Denmark, provided interesting information on food habits during the Iron Age. Their last meals contained 66 different plant taxa (Godwin 1960; King 1966), i.e., a diet much more diversified than that of modern man. According to FAO’s Production Yearbooks, developed countries, and especially urban populations, began to depend almost entirely on the extensive and intensive agricultural production consisting in merely a little more than 100 food plants. Part of the old tradition is still maintained in some Latin American and eastern Asian countries. In Mexico, more than 20 "weeds" are also cultivated (Linares and Aguirre 1992). For instance, in Korean local markets 112 wild plants are sold at prices higher than those of cultivated species. Moreover, eleven species (some of them "weeds") are exported to the U.S.A. and used to prepare Korean and Chinese typical dishes (Pemberton and Lee 1996). Similarly, Moroccan weeds are exported with the same purpose to the U.S.A., Spain, Italy and Greece (Tanji and Nassif 1995).

The use of edible wild plants and weeds has been considered by several authors (Harris 1969, Kunkel 1984, Facciola 1990, Zurlo and Brandão 1990, Duke 1992, Linares and Aguirre, eds. 1992, and references there). Some of these books only provide extensive lists of species, with indications of their edible parts, while others add descriptions and illustrations of the species. Clarke (1977), Michael (1980), and Linares and Aguirre (1992) include numerous recipes. Edible weeds, however, are scarcely used in many countries, and weed gathering is more of a weekend hobby than a regular source of food supply.

To our knowledge, no attempts have been made to assess quantitatively the potential amount of food provided by common weeds and escapes. ]]> Our interest was to evaluate the available biomass of human food provided by weeds in urban, periurban and field habitats in the northern and southern Neotropical region. We tried to compare two samples obtained in different climatic regions, i.e., tropical (Coatepec, Mexico) and temperate to cold-temperate (Bariloche, Argentina) areas.

The environments selected were abandoned fields, main (paved) roads, secondary roads (suburban dirt roads), pathways, vacant urban and suburban lots. Sugar cane fields, milpas, coffee plantations (only Coatepec), and orchards (only Bariloche) were also considered. In these cases, we gathered and weighed all the edible parts of species appearing in 0.25 m² quadrats (50 x 50 wooden frames). Along roadsides (haphazardly selected), we sampled at repeated 1 km distances, by laying ten quadrats regularly at 2 m distance each, parallel to the pavement. Urban vacant lots and fields were randomly sampled. The frames were laid along a random walk using a stopwatch to select compass direction and number of steps. The collected plants of each sample were kept in plastic bags and transported to the laboratory, separated by species and their edible portions were weighed with 0.01 g precision. Water content was calculated after 72 hr dehydration at 60-80 º C (Rapoport et al. 1995). Additionally, in some cases, the yields were assessed in terms of biomass collected per minute of harvesting. This survey was performed in order to evaluate whether there is any profit in gathering a given species in sites where it is abundant. In this way, we were in the position of a person searching for food. The analysis of the proportions of edible parts (roots, leaves, fruits, seeds) of all Mexican and Argentine weed species was obtained from Kunkel’s (1984) list, and from our own records. Comparisons of fresh edible weights among habitats were made by means of Kruskal-Wallis tests. The comparison of total fresh weights between Mexico and Argentina was made by means of a Mann-Whitney test. In the case of significant differences, the Student-Newman-Keuls’ test of multiple comparisons was applied. The detailed statistical analysis of data is reported elsewhere (Ladio et al. 1998).

The main roads sampled in Bariloche were roads No. 258 (Bariloche-El Bolsón), No. 237 (Bariloche-Alicura and Bariloche-Airport) and No. 237 (Bariloche-Llao Llao). In Coatepec, the road sampled was Carretera Briones.

 
Climate

Comparative data on climates appear in Table 1.

The majority of the sampled species appear in Kunkel’s (1984) list of food plants, to which we added Osmorhiza chilensis, from Argentina, and Drymaria gracilis, Galinsoga quadriradiata, Hydrocotyle bonariensis, Hydrocotyle mexicana, Margaranthus sulphureus, Oxalis latifolia, Sida glabra, S. spinosa, and Tripogandra, serrulata from Mexico. These plants are commonly consumed by people and were repeatedly tasted by us.

In total, 43 species were recorded in Coatepec and 32 species in the Bariloche samples (Tables 2 and 3). They represent, however, a small fraction of the real richness in both areas. More than 24% of the 700 weeds listed in Mexico by anonymus (1991) are edible. Of the 320 exotic weeds recorded in northwestern Patagonia (Rapoport and Brión 1991), 90 species (28%) are edible. The Catalogue of Mexican Weeds (anonymus 1991) includes 168 edible species of which 36.3% are perennial and 63.7% annual and/or biennial. Similarly, in Argentina, the proportions are 35.0% and 65.0%, respectively, from a total of 160 edible weeds that are mentioned in Marzocca et al. (1976). Perennials show a significantly higher proportion of edible roots (including tubers or rhizomes) and herbals than among the annual-biennial species (P = 0.027 and 0.021, respectively; Table 4). In contrast, annual-biennials show a greater proportion of edible leaves and seeds than perennials (P = 0.008 and 0.012, respectively) as showed by means of a 2 x 2 contingency table.

  ]]>  
 

Yields per hectare: Because the number of sites surveyed and the number of samples per site were numerous, we made a tentative assessment of the amount of food available per hectare (Table 5). In the case of plantations and orchards, cultivated plants were not included. Coatepecan yields (2.1 tons, on average, per hectare) were apparently higher than the Barilochean yields (1. 3 tons per hectare). The extreme maximum yield was recorded on a main road between Coatepec and Xalapa (10,158.0 kg/ha) based on 10 samples taken in January 1996, and on a vacant lot in Bariloche (7,542.9 kg/ha) based on 7 samples taken in December 1996. When comparing equivalent Coatepecan and Barilochean environments by means of a Mann-Whitney test, the differences were not significant (P>0.05). However, the differences in the median values among total yields per hectare of Mexico and Argentina were significant (U = 880; P = 0.03)
 

Gathering harvest. Species’ yields: The following data correspond to fresh weights of the edible parts obtained per species in a given period of time. Sites were selected for their special abundances, from the point of view of a person searching for food. The process of gathering in Mexico took 15 minutes for each sample (five samples per species) and the mean values are presented per minute (Table 6). ]]>  

In Bariloche, only three species were evaluated. (i) Taraxacum officinale (8 sampling sites, 21 minutes total gathering time). Depending on its abundances, yields varied between 9.1 and 36.4 g/min, with a mean of 17.0 ± 8.8 g/min. (SD) of leaf blades without their central nerves. (ii) Chenopodium album (3 samples, 9 minutes gathering time). Yields were from 38.3 to 56.3 g/min, with a mean of 48.0 ± 9.1 g/min. Only leaves and tender apical stems were considered. (iii) Montia perfoliata (11 sampling sites, 40 minutes total gathering time). Yields from 31.0 to 239.6 g/min, with a mean of 80.4 ± 65.0 g/min. Leaving aside the figures obtained in Heliconia caribaea, which represents a special case for its edible roots, yields seem more productive in Bariloche than in Coatepec. Outstanding biomasses are produced by Montia perfoliata, a North American invader of Patagonian urban forests.

The most profitable plants in Bariloche: Montia perfoliata (=Claytonia perfoliata)

This species shows clear capabilities to recover after harvesting. During the 1995 growth season, in Bariloche we measured the fresh biomasses of three 0.25 m² plots completely dominated by M. perfoliata. The procedure was repeated on the same plot ca. 30 and 60 days later, with the following results (in grams).

 
On the basis of 34 samples where M. perfoliata was abundant, food biomasses varied from 108.5 to 973.6 g per 0.25 m² plot. Mean value 275.9 ± 210.6 g (SD). The aerial parts of this plant are 100% edible. Average water content varied around 78.5 ± 7.3 %.
 

Plantago lanceolata
Leaf fresh weight per 0.25 m² plots (n = 25 plots) showed figures ranging from 39.2 to 309.2 g. Mean value 124.1 ± 60.2 g. Seeds should be added to this food source. At the end of the summer, we selected three 1 m² plots showing ripe fruits. On average, there were 356.0 ± 38.4 spikes/m² in high density patches. Each spike may contain about 500 seeds measuring 2 mm each, showing about 36 % abortions. Viable seeds were assessed in 162.1 g/m². Although the process is laborious (about one hour-person to screen the seeds of one square meter), it is useful to know that a one hectare monospecific plot may yield more than 1.5 tons of seeds, in addition to almost 5 tons of leaves. Of course, it would imply the design of adequate mechanical technology to accelerate the process of screening seeds. Average water content of leaves varied around 78.6 ± 4.9 %. ]]>  

Rumex acetosella
Yields vary between 8.9 and 186.7 g/0.25 m². Mean 53.3 ± 47.9 g (n = 15 samples). The species is widely distributed in disturbed and undisturbed habitats, especially in surrounding grasslands and pasturelands. Average water content varied around 77.1 ± 4.6 %.

Taraxacum officinale
Yields vary between 11.2 and 107.4 g/0.25 m² (central ribs excluded). Mean 47.2 ± 27.6 g (n = 17 samples). Average water content varied around 82.2 ± 4.7 %.
 

Common weeds prove to be an interesting resource in small to medium-sized human settlements where they may provide supplementary food. In large cities, suburban populations may also profit from edible weeds. The data from Table 7 show that edible, non-weedy plants comprise between 6 and 21 percent of the biomass of the natural communities surveyed. The proportion of edible plants increases considerably in anthropic environments, especially in weed communities. Roughly, ten percent of the 260,000 known vascular species of the world should be considered as a potential source for human consumption. It is probably not by mere chance that the majority of the centers of origin or domestication of cultivated plants proposed by Vavilov (1938) corresponded to ancient, sedentary cultures. The idea that civilizations arose in areas with abundant edible plants adequate for culturing is probably incorrect. Since edible plants abound everywhere, it seems that civilizations developed in any environment where for historical reasons, people had time enough to exploit their natural resources in a more permanent and intensive way. By selection, plants originally wild, began to be more productive and adequate for human consumption. Rye, oats, carrots and several other cultivated plants originated as weeds, a fact that gives a clear idea of the enormous potential of weeds and other wild non-weedy plants as a source for new cultures. The economic incentive provided by the revival of ancient gastronomic traditions persuaded some private entrepreneurs to change from the occasional gathering to a more permanent cultivation of "weeds". Popular markets in Mexico (Linares and Aguirre 1992) and Korea (Pemberton and Lee 1990) offer a variety of gourmet "weeds" at higher prices than the common vegetables.

A case of human-livestock-plant mutualism is mentioned by Kuznar (1993), especially referred to species of Chenopodium proliferating in unusual concentrations in corrals. Herd animals transport these forage species to pastoral campsites where the plants thrive in the organic corral soils. This creates a mutually beneficial relationship where certain plant species become camp followers of pastoral campsites. This is the process by which plant invaders reach the status of weeds first and, later on, the status of cultivars, according to Vavilov (1938). And this process may explain the fact that the majority of the most aggressive and cosmopolitan weeds are edible. Initially, plant domestication began early, and in an unconscious way, probably in the Paleolithic Age, in primitive hunter-gatherer cultures (Rapoport et al. 1995). It is interesting to note that leaf-cutting ants (Acromyrmex), as determined by Farji-Brener (1996), show clear foraging preferences for exotic ruderal weeds which, according to Coley et al. (1985), are plants that invest more energy in reproduction, growth and dispersal than in anti-herbivore chemical defenses. It is suggestive that 52% of the early introductions (archaeophytes) in Poland listed by Trzcinska (1982) are edible. ]]> Our results show that in anthropic habitats there are immense amounts of edible plants which are not always totally profited from. This is clearly evident in Argentina where people have almost lost the ancient practice of gathering wild food plants. In a tropical area such as Coatepec, the overall ‘standing crop’ averages 2.1 tons/ha whilst in temperate Bariloche it reaches 1.3 tons/ha. Tropical weeds are richer in species number and more productive than temperate weeds. The latter are almost absent during the long winter season, although they can be dehydrated and cooked during the cold season.

Because a significant sector of the Argentine population suffers from serious problems of malnutrition, the Universidad del Comahue and Municipalidad de Bariloche published a booklet (Rapoport et al. 1997) illustrating the 20 most common edible weeds. Free copies were distributed in provincial public schools. This instruction manual represents the beginning of a program which hopes to restore, at least partially, our ancestors’ knowledge. As a result of this printed information and a television program, a substantial increment of popular awareness and utilization of this resource, was registered.
 
 
Acknowledgements

Our thanks to Andrea Premoli, Tom Witham and Barbara Drausal for their useful comments on the article.
 
 Resumen

Las malezas abundan en ambientes urbanos y rurales. Según la región y lugar, hasta el 66% de las especies de malezas pueden ser comestibles y constituir un recurso alimentario adicional para el ser humano. Sobre la base de 400 muestras de ¼ m² cada una, recolectadas en áreas tropicales (rutas, terrenos baldíos, calles y plantaciones en Coatepec, México) el promedio de la biomasa en peso fresco varió entre 1 277 y 3 582 kg/ha. Un muestreo similar en un área templada (739 muestras en Bariloche, Argentina) arrojó valores medios entre 287 y 2 939 kg/ha. En total se registraron 43 especies en Coatepec y 32 especies en Bariloche. La media general (total) fue de 2.1 y 1.3 ton/ha, respectivamente. A una escala geográfica mayor, una comparación entre las malezas mexicanas y argentinas no mostró mayores variaciones regionales en cuanto a qué partes u órganos son los comestibles. En ambos lugares, el orden de aprovechamiento, de mayor a menor, fue: hojas, semillas, raíces, frutos, infusiones, flores y condimentos. Las raíces comestibles (incluyendo bulbos y rizomas) parecen ser más comunes entre las especies perennes que entre las anuales.
  ]]> References

Anonymous. 1991. Catálogo de especies vegetales consideradas malezas. Asoc. Mex. Ciencia de la Maleza, AMCM México, D.F. p. 92.         [ Links ]

Bianco, C.A. & J.J. Cantero. 1992. Las plantas vasculares del suroeste de la Provincia de Córdoba. Iconografía. Facultad Agronomía y Veterinaria. Universidad Nacional Río Cuarto, Río Cuarto, Argentina. p. 229         [ Links ]

Boom, B. M. 1987. Ethnobotany of the Chácobo Indians, Beni, Bolivia. Adv. Econ. Bot. 4: 1-68.         [ Links ]

Clarke, C.B. 1977. Edible and useful plants of California. University of California, Berkeley, California. p. 280.         [ Links ]

Coley, P.D., J.P. Briant & F.S. Chapin. 1985. Resource availability and plant anti-herbivore defense. Science 230: 895-899.         [ Links ]

Derksen, D. A., A. G. Thomas, G. P Lafond, H. A Loeppky & C. J. Swanton. 1993. Impact of agronomic practices on weed communities: Fallow within tillage systems. Weed Sci. 42:184-194.         [ Links ]

Domínguez Barradas, B. A. 1995. La vegetación vascular espontánea en calles y banquetas de la Ciudad de Xalapa, Veracruz. Tesis Licenciatura, Universidad Veracruzana, Facultad de Biología, Veracruz, México. p. 93.         [ Links ]

Duke, J. A. 1992. Handbook of edible weeds. CRC, Boca Raton, Florida. p. 246.         [ Links ]

Facciola, S. 1990. Cornucopia. A source book of edible plants. Kampong, Vista, California. p. 678.         [ Links ]

FAO. 1987. Production yearbook. Food and Agriculture Organization, United Nations, Rome: 1-306.         [ Links ]

Farji-Brener, A.G. 1996. Posibles vías de expansión de la hormiga cortadora de hojas Acromyrmex lobicornis hacia la Patagonia. Ecología Austral 6: 144-150.         [ Links ]

Felger, R.S., and G.P. Nabhan. 1978. Agroecosystem diversity: A model from the Sonoran desert. AAAS (Am. Assoc. Adv. Sci.) Selected Symposia 10.         [ Links ]

Franceschi, E. A. 1996. The ruderal vegetation of Rosario City, Argentina. Landscape Urb. Plann. 34: 11-18.         [ Links ]

Frankton, C. & G.A. Mulligan. 1977. Weeds of Canada. Canada Deptartment of . Agriculture, Publication 948. p. 217.         [ Links ]

Godwin, H. 1969. The history of weeds in Britain. p. 1-10. In: The biology of weeds. J.L. Harper (ed.). Blackwell, Oxford, U.K.         [ Links ]

Gómez, C.M. & M. Soto. 1990. Climático del Municipio de Coatepec. Instituto de Ecología, Publ. No. 27, Xalapa, Veracruz, México.         [ Links ]

Grigera, D., E. Bianchi, C. Brión, J. Puntieri & N. Rodríguez. 1989. Carta del medio ambiente y su dinámica de San Carlos de Bariloche, 1987. Universidad Nacional del Comahue, Bariloche, Argentina. p. 40.         [ Links ]

Harris, B.C. 1969. Eat the weeds. Keats Publishing Inc., New Canaan, Connecticut, p. 254.         [ Links ]

Holm, L.G., D.L. Plucknett, J.V. Pancho & J.P. Herberger. 1977. The world’s worst weeds. University of Hawaii, Honolulu, Hawaii, p. 609.         [ Links ]

King, L. W. 1966. Weeds of the world. Biology and Control. Leonard Hill, London. p. 526.         [ Links ]

Kunkel, G. 1984. Plants for human consumption. Koeltz Scientific Books, Koenigsten, Germany. p. 393.         [ Links ]

Kuznar, L.A. 1993. Mutualism between Chenopodium, herd animals, and herders in the south central Andes. Moun. Res. Dev. 13: 257-265.         [ Links ]

Ladio, A.H., E. Raffaele, L. Ghermandi, M.E. Díaz-Betancourt, I. López-Moreno & E.H. Rapoport. 1998. Las malezas en las ciudades de S.C. de Bariloche y Coatepec y su potencial oferta alimentaria. Actas XIII Congr. Latinoam. Malezas, Buenos Aires, Argentina.         [ Links ]

Lema, O. W. 1988. Catálogo ilustrado de la flora de Salto Grande. Introducción a su estudio. Tomo I. Comisión Técnica Mixta de Salto Grande, Argentina-Uruguay. p. 312.         [ Links ]

Linares, E. & J. Aguirre (eds.) 1992. Los quelites, un tesoro culinario. Universidad Autónoma de México & Instituto Nacional de la Nutrición "Salvador Zubirán", México. p. 141.         [ Links ]

Michael, P. 1980. All good things around us. A cookbook and guide to wild plants and herbs. Holt, Rinehart & Winston, New York, p. 240.         [ Links ]

Martínez-Crovetto, R. 1968. Estudios etnobotánicos. IV. Nombres de plantas y su utilidad, según los indios onas de Tierra del Fuego. Etnobiologica 3: 12-19.         [ Links ]

Marzocca, A., O.J.V. Mársico & O. Del Puerto. 1976. Manual de malezas. Hemisferio Sur, Buenos Aires, Argentina. p. 564.         [ Links ]

Misra, J., H. N. Pandey, R. S. Tripathi & U. K. Sahoo. 1992. Weed population dynamics under 'jhum' (slash and burn agriculture) and terrace cultivation in northeast India. Agric., Ecosyst. Env. 41: 285-295.         [ Links ]

Pemberton, R. W. & N. S. Lee. 1996. Wild food plants in South Korea: market presence, new crops, and exports to the United States. Econ. Bot. 50: 57-70.         [ Links ]

Peters, C. M., A. H. Gentry & R. O. Mendelsohn. 1989. Valuation of an Amazonian rainforest. Nature 339: 655-656.         [ Links ]

Rapoport, E.H., & C. Brión. 1991. Malezas exóticas y plantas escapadas de cultivo en el noroeste patagónico. Segunda Aproximación. Cuadernos de Alter Natura No. 1, Bariloche, Argentina. p. 19.         [ Links ]

Rapoport, E.H., L.S. Margutti & E.H. Sanz. 1997. Plantas silvestres comestibles de la Patagonia Andina. Parte I. Exóticas. Programa Extensión Universitaria, Univ. Nac. del Comahue, Depto. Ecología, CRUB, Bariloche, Argentina. p. 51.         [ Links ]

Rapoport, E.H., E. Raffaele, L. Ghermandi & L. Margutti. 1995. Edible weeds: A scarcely used resource. Bull. Ecol. Soc. Amer. 76: 163-166.         [ Links ]

Rudloff, W. 1981. World-Climates. Wissenschaftliche Verlagsgesellschaft, Stuttgart, Germany. p. 632.         [ Links ]

Sánchez, P. & H. Huranga. 1990. Malezas importantes de Cuba. Instituto de Investigación Fundamental de Agricultura Tropical A. de Humboldt, Ministerio de Agricultura, Cuba. p. 63.         [ Links ]

Tanji, A. & Nassif, F. 1995. Edible weeds in Morocco. Weed Technol. 9: 617-620.         [ Links ]

Trzcinska-Tacik, H. & K. Wasylikowa. 1982. History of the synanthropic changes of flora and vegetation of Poland. Mem. Zool. 37: 47-69.         [ Links ]

UNC. 1983. Cartas de precipitación de las provincias de Río Negro y Neuquén. Facultad Ciencias Agrarias, Universidad Nacional del Comahue, Cinco Saltos, Argentina.         [ Links ]

Vavilov, N.I. 1938. Los centros genético-geográficos de las plantas cultivadas. Rev. Fac. Agron. Univ. Nac. La Plata 22: 65-94.         [ Links ]

Wilkinson, R.E. & H.E Jaques. 1979. How to know the weeds. W.C. Brown, Dubuque, Iowa p. 230.         [ Links ]

Zurlo, C. & M. Brandão. 1990. As ervas comestíveis. Globo, São Paulo, Brazil. p. 167.         [ Links ]
 

1. Instituto de Ecología, Ap. Postal 63, Xalapa 91000, Veracruz, México. martha @ecología. edu.mx
2. Universidad Nacional del Comahue, Departamento de Ecología, 8400 Bariloche, Argentina and CONICET rapoport@cab.cnea.edu.ar
3. The sequence of authors was alphabetically ordered ]]> 1991 92 1992 229 1987 4 1-68 1977 280 1985 230 895-899 1993 42 184-194 1995 93 1992 246 1990 678 FAO 1987 1-306 1996 6 144-150 1978 1996 34 11-18 1977 948 217 1969 1-10 1990 27 1989 40 1969 254 1977 609 1966 526 1984 393 1993 13 257-265 1998 1988 312 1992 141 1980 240 1968 3 12-19 1976 564 1992 41 285-295 1996 50 57-70 1989 339 655-656 1991 1 19 1997 51 1995 76 163-166 1981 632 1990 63 1995 9 617-620 1982 37 47-69 UNC 1983 1938 22 65-94 1979 230 1990 167