On December 16, 1983, Dr. Norman Borlaugh (with others) visited the National Agrarian University in La Moina, Peru... In their talks they emphasized the development of Opaque Maize and the importance of the fact that its protein content, in the form of lysine... has been increased from 2.8 to 4%... It was pointed out (by the Andeans) that in both centers of pre-Hispanic maize cultivation, the Central Andes and Meso-America, indigenous farmers intuitively knew that maize's nutritional imbalance could be improved, not by forcing a plant to produce a balanced food, but by associating it with another crop that had a high content of the amino acids that maize was lacking... In the Andes maize was associated with quinoa... while on the Mexican plateau its relative Huazontle (Chenopodium Nutalliae) was cultivated.
Editorial of the Andean Food Crops Bulletin, March 1984
Most of the Indian products are still under cultivation....On the plateau (Peru) the grain crops were quinoa and canuagua. The quinoa struck me as being a crop from which a choice breakfast food could be made. It was starchy with just enough tang from its pigweed blood to produce a good flavor.
H.V. Harlan's One Man's Life with Barley (1953)
Quinoa has a long and distinguished history in South America. Quinoa has been cultivated in the Andean highlands since 3,000 BC (Tapia 1982). In the Quechua language of the Incas, quinoa is the chisiya mama or "mother grain;" in Spanish, it is quinoa, trigo inca, or arroz del Peru (National Research Council 1989). Its adaptation to cold, dry climates, seed processing similarity to rice, and excellent nutritional qualities make quinoa a crop of considerable value to highland areas around the world which are currently limited as far as crop diversity and nutritional value. Development of other Chenopodium species in the United States (Wilson 1981) the Himalayas (Partap and Kapoor 1985), Mexico (Risi and Galway 1989), and Denmark (Renfrew 1973) illustrate a diverse appreciation for this genera. Even the weedy relatives, C. alba and C. berlanderii, in the United States and around the world have been utilized as a food during times of starvation.
The nutritional value of quinoa has been known for a long time to be superior to traditional cereals and is, in fact, superior to milk solids in feeding trails (White et al. 1955). Protein content ranges from 10 to 18% with a fat content of 4.1 to 8.8%. Starch, ash, and crude fiber average 60.1, 4.2, and 3.4%, respectively (DeBruin 1964; Ballon pers. commun.). The ash has been found to primarily consist of potassium and phosphorus (65% of total). Calcium and iron are significantly higher in quinoa than in rice, maize, wheat, or oats (White et al. 1955; DeBruin 1964). Variations have been observed between species and between landraces within species. Many landraces of quinoa contain saponin in the seedcoat. Saponins function as "antinutrients" and are frequently associated with plant lipids. They are not normally absorbed from the gut and have been shown to induce small intestinal damage or reduce intestinal absorption of nutrients (Jenkins 1988). Quinoa saponin is a known hemolytic when mixed with blood cells. In South America, saponin removed from quinoa is used as a detergent for clothing, washing and as an antiseptic to promote healing of skin injuries (D. Cusack pers. commun.; E. Ballon pers. commun.). Saponin can be removed either mechanically or with a water rinse (White et al. 1955; DeBruin 1964; Mahoney et al. 1975). Mechanical abrasion systems currently in use fail to remove all saponin, leaving bran with saponin attached to perisperm granules (Becker and Hanners 1991).
Mechanical abrasion of quinoa has been found to significantly increase a-amylase activity in quinoa (Lorenz and Nyanzi 1989). This appears to be due to the reduction in the pericarp content during abrasion. Quinoa has a significantly higher alpha-amylase content than cereals such as rice and proso millet. Amylograph analyses also show quinoa to have superior alpha-amylase activity to wheat (Lorenz and Nyanzi 1989). The mineral content of quinoa has been evaluated both in South America by Ballon (pers. commun.) and by the author for North American quinoa. Results indicate a mineral profile which is generally higher than comparable cereals (Table 1).
Quinoa has received a considerable amount of attention in the Andean highlands over the past two decades (Johnson 1990). Quinoa production extends from Columbia to Chile and Argentina and a diversity of landraces have resulted. Within the major quinoa production areas of Columbia, Ecuador, Peru, and Bolivia, Gandarillas (1968) described 17 races based upon morphological characters. Galwey (1989) and Tapia (1979) proposed four main types based upon geographic location: The "Valley" type, typical from 2,000 to 4,000 m in elevation; the "Altiplano" type, typical of highland areas above 4,000 m; the "Salar" type of 4,000 m but adapted to the high pH soils typical of the Atacama region; the "Sea Level" types found in the inner valleys of Bolovia. The "Sea Level" type has been described by Wilson's electrophoretic work to be distinctly different from the other highland quinoas. Cusack (1984) proposed that quinoa within South American cultures may have arisen independently. Wilson (1990) has proposed a division of quinoa into two distinct groups: northern and southern. The northern group, being less diverse than the southern, would indicate it is derived from the southern group. In Colorado crossing trials, quinoas within Wilson's groups have shown no heterosis for yield while crosses between his groups have shown heterosis varying between 201 and 491%.
The heterosis demonstrated in quinoa initiated a search for male sterility in 1987 and 1988. Using commercial fields in Colorado, no male steriles were found in the cultivar 'CO 407'--a Southern, Sea Level type of quinoa common to Colorado. In 1989, Ward (1991) obtained two potential sources of male sterile--one from the cultivar 'Amachuma' and another from the cultivar 'Apelawa'. The 'Amachuma' type appears to be a simply inherited, genetic male sterile. The 'Apelawa' type is a cytoplasmic male sterile and Ward (1991) has transferred this trait into four additional background genotypes. Other sources of male sterility in quinoa have been reported such as genic-cytoplasmic male sterility (Simmons 1971); a single gene male sterile (Gandarillas 1979); and a cytoplasmic male sterile reported by Galwey and Risi (1984). Research on male sterility in quinoa has generally not received a great deal of attention or replication and at this point, only Ward's work appears definitive.
Evaluation of germplasm can be difficult given the array of accessions available. One method to separate the diversity modern researchers have observed in quinoa is through canonical discriminate analysis. Risi and Galwey (1989) used a canonical discriminant analysis with 19 traits for accessions from sea level types, altiplano types, salar types, and valley types. Risi and Galwey found the sea level and valley types to be very homogeneous within type. Other types were difficult to discriminate. E. Ballon and D. Johnson (unpubl.) studied 15 cultivars and found altiplano, valley, and sea level types were not homogeneous within type when grown in Colorado and New Mexico.
Significant differences in agronomic value in the United States were observed between 15 cultivars for protein, seed size, and yield, with no significant location effects. There was, however, a significant cultivar by location interaction in all cases (Ballon et al. 1991). Similar yield results were observed by Galwey and Risi (1984) in England and in Finland (Carmen 1984). Ballon et al. (1991) also conducted a heritability for grain yield and grain size. Broad-sense heritabilities of 49 and 32% were estimated for yield and seed size, respectively. Heritability for protein content was not calculated.
Irrigation may have a significant effect on yield. Tapia (1984) recommends an average of 550 mm of available moisture. In the loamy soils typical of the Valley quinoa types, 700 mm may be required while types which grow in the saltflats of Southern Bolivia require 350 mm. In Colorado, Flynn (1990) found maximum yields of 1439 kg/ha were obtained on sandy-loam soils with 208 mm of water (rainfall and irrigation) with available water levels of 128, 208, 307, and 375 mm being tested.
Seeding rates vary between United States recommendations and those of South America. Current United States recommendations are for seeding rates of 1 to 1.5 million plant/ha (Johnson and Croissant 1985). South American recommendations are for 8 million/ha for row cropping and 20 million/ha for broadcast cultural practices.
Weed control has had a major impact on quinoa yield. In Colorado, grassy weed control alone increased yields from 640 kg/ha to 1,822 kg/ha (Johnson 1990). Weed control via herbicides have been effective and several show promise. In England, Metamazide, Propachlor, Linuron, Propyzamide, and aloxium sodium did not significantly reduce plant stands of two quinoa cultivars (Galwey and Risi 1984). In Colorado, preliminary herbicide studies of pre-emerge herbicides with Dual, Furloe, Sutan, and Antor showed good crop safety and control of grasses and many broadleaf weeds (Westra 1988). Post-emergent control was best for Poast, Tough, and Probe, with Tough and Probe at lower rates (Westra 1988).
Salt tolerance is of concern to quinoa growers in the arid regions of the world. Preliminary work by the University of Arizona indicates that quinoa is very salt tolerant. Germination of five quinoa cultivars was unaffected by salinity levels varying from 114 to 2,169 ppm NaCl. Growth over a 6-week period shows CO 407 to be superior to other cultivars at the 2,169 ppm level with only a 40% reduction in plant height. 'Isluge' (CO 211), a salt adapted cultivar, had a 45% reduction in plant height (D. Shropshire and V. Lindley pers. commun.).
Insects are a major concern in South America and have received increasing attention in the United States. In South America, Romero (1980) has pointed out two important pests of quinoa: Kcona Kcona (Scrobipalpula sp.) which destroy buds, inflorescences, immature and mature grain; and "leaf miners" (Liriomyza sp.) which destroy leaves and occasionally stacks of quinoa. In Colorado, Cranshaw et al. (1990) found insects commonly associated with sugar beet and lambsquarters (C. alba L.). Seedling damage was caused by Malanotrichus coagulatus Ulher and Atomoscelis modestus Van Duzee and the seed bug, Nysius raphanus Howard. Foliar pests include two leaf miners, Pegomyia hyoscyami Panzer and Monoxia nr. pallida Blake. Also found were leaf feeding insects such as a leaf curling aphid Hayhursita atriplicis (L.) and various Lepidoptera, such as Spodoptera exiqua Hubner. A foot, or root feeding aphid, Pemphgus populivenae Fitch caused late season damage. Seed damage was due to Lygus spp.
There are apparently two types of saponin: (1) a rarer acid and neutral saponin group (found in white quinoas) that can be used commercially in the production of pharmaceutical steroids, and (2) a more common type prevalent in the yellow quinoa cultivars which is used in soaps, detergents, beer production, fire extinguishers, photography, shampoos, cosmetics, and pharmeceuticals (synthetic hormones). Within the saponin of quinoa, two aglycones predominate: oleanolic acid and heterogenin. The ratio of oleanolic acid to heterogenin varies from 2.3 to 8.6 (Burnouf-Radosevich et al., 1983) depending upon the cultivar. Oleanolic acid and heterogenin content averaged 3.0 and 1.4 mg/g. Triterpene concentration ranged from 6.3 mg/g for 'Real de Puno' to as low as 0.06 mg/g for 'Sajama' and was found to be highly proportional to saponin content.
Once saponin is removed, protein quality was unaffected. Amino acid balance was virtually the same regardless of saponin content of the seed (Burnouf-Radossevich et al., 1983). Both soluble proteins (albumins and globulins) and the high molecular weight glutelins were measured. High saponin cultivars lacked three glutelin subunits and may be useful as protein markers for low-saponin type quinoas.
Processing quinoa to remove saponin can be done by alkaline water washing or mechanically via abrasion. Mechanical dehulling involves "pearling" the grain to remove the pericarp as bran. High saponin cultivars require more abrasion than low saponin types. Washing quinoa is probably not at the current time a viable option in developed countries due to pollution of water where potential pollution of major bodies of water may be affected. In South America, a "dry" system involving a flotation cell is used. The seed is wetted and dirt, saponin, and foreign matter are removed and the grain then subjected to forced air to dry. Foam breakdown and drying costs are serious problems in this method. The grain can also be "pre-toasted" and "polished" or abraided using a spinning stone (Gandarillas 1982). Abrasion also tends to reduce ash content in quinoa and has been demonstrated to increase alpha-amylase activity. Reichert et al. (1986a,b) found the saponin content of quinoa flour could be effectively reduced by minimal abrasion in a Tangential Abrasive Dehulling Device (TADD).
Becker and Hanners (1991) used three quinoas (low, medium, and high saponin content) and a Morehouse Model 350 stone mill to evaluate saponin removal. The stone mill removed 33 to 40% of the seed as bran fraction. Moisture tempering the grain from 8 to 16% did not improve milling yield or mill fraction composition. The mill fraction typically contained less than 0.3 mg/g of saponin and residual bran was left on the grain. The extremely high loss of grain via mechanical milling indicates an area of research which needs more work.
Nutritional analyses of quinoa have generally been conducted on desaponified quinoa. In one such test, Dahlin (1991) tested quinoa against high and low tannin sorghum, corn, wheat, rye, and proso millet. In the unprocessed (nonextruded) state, quinoa was found to be the most carbohydrate digestible of the seven crops. Dahlin states that this is probably due to the very high concentration of alpha-amylase in quinoa. Quinoa carbohydrates improved in digestibility when extruded at 25% feed moisture and a 100°/150°C initial/final food temperature, but unlike cereals, overall digestibility decreased with extrusion. Dahlin measured nitrogen solubility, a measure of biological value of protein. Nitrogen solubility was highest for quinoa and rye and ranged from 20.9 at pH 2.0 to 30.0 at pH 10.
Kleiman et al.(1972) evaluated oil from four species of the Chenopodiaceae and found most contained some unusual components. These are methyl cis-5-hexadecenoate (4.6 to 12%) and methyl 5-octadecenoate (1.1 to 1.2%). Kleiman et al. also observed quinoa oil to contain small amounts of 18:25,9 and 18:35,9,12. The majority of the oil, however was common to most seeds with oleic fatty acids composing 14 to 21% of the oil. Linoleic and linolenic fatty acids composed 53 to 57% and 3.5 to 7.8% respectively. These oil compositions are similar to those obtained from quinoa's distant relatives, spinach (Spinacea oleracea L.) and Russian thistle (Salsola pestifer).
The unusual components, methyl cis-5-hexadecenoate and methyl 5-octadecenoate occur in other plant species but are the highest in concentration in the Chenopodiaceae. Quinoa was unique among the four species in that it lacked methyl cis-5-hexadecenoate and methyl 5-octadecenoate in detectable amounts. Quinoa oil was composed of 31% oleic; 45% linoleic; and 2.7% linolenic fatty acids. The remainder of the oil was composed of 16, 20, and 22 carbon fatty acids.