A survey was undertaken in order to understand the social, technological, economic and potential use of the "nopal" vegetable (Oputia ficus-indica) in the region of Xochimilco. The results showed that nopal is grown either within the producer's home garden or on land near the town. There is no defined season of planting a new crop, though nopal production is closely associated with seasonal patterns. The ratios of using organic to inorganic fertilizers were 3:7, 6:1 and 6:0 on small, medium and large plots respectively. Insects were the only problem affecting yield. Experimental micro-silages were set up using nopal to assess the nutritional content as food for ruminants. The addition of molasses and maize stover to the micro-silages of nopal reduced crude protein content and increased DM.

The importance of nopal vegetable production systems is discussed in terms of a suitable crop as a sustainable model of production in the region.

The nopal-vegetable (Opuntia ficus-indica), one of the many classified Mexican agaves (Sanchez and Castañeda 1991) is extensively used in the region of Xochimilco as a natural perennial resource to feed and provide economic income for the local population. The origin of nopal cultivation dates to prehispanic times, the methods of production in many ways being the same today as they were before the conquest, for example cultivation on terraces. The consumption of nopal leaves is an important cultural component of the regional cuisine in the Mexican Basin (Rojas 1990). According to the data, nopal production in the region covers an area of 7000 hectares with an estimated annual production of 90,000 tons of leaves (COCODER 1985), 50% of which is consumed within the City of Mexico whereas the rest is distributed to neighbouring states and a small percentage exported to the USA and Japan. The main characteristics of the edible cactus are good adaptation to poor soils, low requirements of water and technology, an intensive use of manure and organic matter to maintain growth and the potential to produce one crop per week which, within sustainability proposals, makes the system an interesting subject to research (Preston and Murgueitio 1992). Although nopal has been widely investigated by many researchers (Thissenhusen 1976; Delgado 1985; Pimienta 1993) to date integrated information of the whole system is scarce.

The objective of the study was to characterise the production of the nopal-vegetable in the terrace areas of the region of Xochimilco and evaluate its potential for ruminants when ensiled.

Physical features of the area include an altitude ranging between 2470 to 2750 masl and a topography within the ranges of 20% to 60% of gradient. Soil origin is volcanic with the presence of riolites, endosites, basalts, lavas, conglomerates, pumice and volcanic ashes. Although the area has some deep soils (> 50 cm) most are within the range of 25 to 50 cm. The zone has no natural rivers but seasonal streams. The climate has been classified as: C (WO) (W) which is sub-humid temperate with a low degree of humidity (Garcia 1973). Mean annual temperature is 16 ºC with an average annual rainfall of 600 to 800 mm per year in the period from May to October. Frosts only occur on the very highest slopes occasionally during the months of December and January. Natural vegetation in the wooded areas is formed by various Pinus species, encinos, sabinos and others. The ground flora is formed by annual and perennial grasses (Graminae) of the predominant genera: Muhlenbergia, Festuca and Sporobolus (Inegi 1994).

The procedure applied to the study included: (i) a survey directed to nopal producers and; (ii) an experimental evaluation to appraise the silage potential of nopal as feed for ruminants. The survey procedure included a questionnaire designed to understand the social, technological and economic aspects of nopal production and was tested successfully in field conditions before its final application. The nopal-vegetable producers were randomly located and the sample included the most important town for nopal production (Villa Milpa Alta). The procedure additionally contained semi-structured interviews at outlets where nopal is prepared and sold at the most influential market in the area, that of Xochimilco. In total, 30 questionnaires were applied.

In the second part of the research, experimental micro-silages of nopal were made to compare the incorporation of two levels of molasses and one micro silage without (0, 7 and 15%) and two levels of maize stover (30 and 35%) in a factorial arrangement (3 x 2) with three replicates. Mature fleshy leaves were used and the micro-silages prepared in 3.5 litre bottles with Bunsen valves. Nopal and maize stover were chopped to a mean size of 2cm and thereafter homogenized with molasses. The ensiling period was 60 days at the end of which the material was analysed for pH, DM, NH₃, N and ash following procedures reported by Tejada (1983). The data obtained in both procedures were analysed using statistical software (SAS 1988) and the results expressed as percentages and means following standard procedures (Daniel 1992).

The production of nopal in the area is situated in two locations, either in a home garden and/or as a unit of production outside town. Family members occupying the household were in the ranges of 2-8 persons. Most of the houses of the producers have services including: electricity (100%), drinking water (100%), drainage (20%) and street pavement (50%). A significant proportion of growers were reported to be illiterate (57%). Levels of education were primary (29%) and secondary (14%). Most of the work with nopal is carried out by men. However, women and children play an important role in the husbandry of associated cultivations as well as in harvesting the leaves and selling the produce in local markets. Work activities of men included: agriculture (70%), skilled work (16%), commerce (7%) and intermediary selling of the produce (7%). Five percent of male producers reported belonging to a society representing nopal growers.

There is no defined season for establishing a new plot of nopal but some distinctions that relate to time of planting other than to the plant itself refer to: money, land availability and a suitable place to plant the crop. Most of the producers interviewed preferred the traditional system of establishing a new growing area (95%), with only a small proportion (5%) showing preferences for a semi-technified system. In both systems planting does not require the furrowing of the new land. Nopal is often planted across the slope of the plots to reduce soil erosion. Distance between plants within the row in most cases was of the order of 20 cm (60%) with a reported range of 10 to 30 cm. Distances between rows (allowing for the presence of a corridor to permit access for harvesting) was 1.0 m (33%) to 1.2 m (60%). Plantlets are selected as mature fleshy leaves with good characteristics of size and vigour then placed in the ground by burying half the leaf in the soil. Number of plants per m² was in the range of 6 to 12 mature plants with most plantings (up to 70%) in the range of 8 to 10 plants per m². The size of the plots in the studied zone (usually measured in terms of "yokes"), is classified between small (0.3 ha). medium (0.6 ha) and large (0.9 ha) which in this study represented 44% of producers for the small, 33% the medium whereas the large plot accounted for 23%.

Usual labours on nopal terraces include those referring to combatting weeds, pruning, fertilization and the control of insects and fungi affecting productivity.

Table 1: Types of fertilizers and amounts used according to the traditional areas (1yoke =0.3ha).
	Inorganic#, kg	Manure, tonnes
Small yoke	750	>200
Medium yoke	850	>300
Large yoke	1000	>400

# Triple 17 (N, P, K) and/or Ammonium sulphate

Weeding is usually undertaken by manual labour with the use of a machete and/or hoe. Trimming is common management carried out as a method to promote regrowth of new tender leaves. This management has a defined seasonal pattern done at the end of the dry season as most producers tend to stop cropping during this time of the year because of a high production of leaves and a corresponding fall in market prices. Similar to weeding, trimming is a manual labour usually made with a machete.

Table 2: Numbers of people for combating weeds, trimming and fertilization in the terraced areas of Xochimilco
	Weeding	Triming	Fertilization
Small yoke	2-3	2-3	3-5
Medium yoke	3-4	3-4	4-6
Large yoke	4-5	4-5	7-8

Table 3: Frequency of common insect problems reported by nopal-vegetable producers in the region of Xochimilco.
Common name	Frequency, %
Blasting worm	19
Red worm	16
"Chahuistle"	14
Spiders	14
Red mosquito	12
Cochineal	9
Louse	6
"Pinacate"	6
Moth	2
Plant louse	2

An important activity refers to fertilization. The amounts and types used are in Table 1. The ratios of organic to inorganic compounds for use on traditional plots were 266:1, 352:1 and 400:1 for small, medium and large, respectively. A possible explanation for such differences may be related to the roles of manuring. Although the main purpose of adding manure is as a source of nitrogen and organic matter, some researchers have suggested an extra role as a heat retainer during cold periods in the months of December and January. This extra function could explain higher use of manure in larger yokes with a higher number of plants. The range in numbers of people used for weeding, trimming and fertilization on traditional plots is presented in Table 2.

As most of the labour is carried out manually, there is an extensive use of people in all activities. However, work attributed to fertilization, which requires distribution of the product over a wide area either at the trunk of nopal (inorganic compound) or all over the field (manure), occupies a higher number of people.

The presence of insect infestation is the most important factor affecting productivity. Frequency of infestation is presented in Table 3. The most important species of insects in the table belong to the first five categories which appear to affect most of the plots whereas the rest were occasional. Many of the infestations are controlled by the use of commercial insecticides including Folidol, Sopracil and Fanela. The utilization of home made insecticides was minimal, only for controlling "chahuistle". Problems with cochineal were managed by scraping the surface of the leaves to remove the pest.

The employment of technology, measured by the utilisation of tools for planting nopal, and the general management and cropping of the agave is presented in Table 4.

Table 4: Implements and the frequency of utilization by producers on nopal-vegetable plots in terrace areas
Implements use by producers	Frequency
Hoe	15
Tlalacho (a type of wide pick axe)	8
Knife	4
Wheelbarrow	3
Rake	2
Leather gloves	2
Large basket	2
Sickle	2
Winnowing fork	1
Pauader	1
Tractor	1
Shovel	1
Pick axe	1
Machete	1
Yoke	1

As can be seen, there is a broad use of manual tools whereas the employment of machinery is minimum. This situation is not rare for the zone as the terrace sizes restrict considerably the use of tractors a fact we have referred to before (Soriano et al 1993). Main tools were the hoe, "talacho" (knife) and wheelbarrow which appear to be frequently used.

Most nopal is grown with few plant associations probably because other species are considered competitors for nutrients and because of the difficult nature of managing additional crops in the rows. However, the plants we have found to be combined with nopal include: the geranium flower (Geranius spp) managed by women and sold as small bunches of flowers in local markets to earn extra income for the family economy. Other species were: lettuce, "quelite", cabbage, carrot, beetroot and radish. Additionally, to increase productivity, producers plant short cycle cultivars in those places where old nopal growing sites are about to be replaced by new plantlets. In this respect the main species were: lettuce (50%), maize/bean/pumpkin complex (20%), spinach (10%) "huauhzontle" (a type of amarantus, 10%) and carrot or green beans (10%). The presence of fruit trees is usual surrounding nopal terraces and include: peach, apricot, lemon, apple and fig.

The mean size of the nopal-vegetable leaf is 20 x 12 cm which is considered first class and is very tender. The larger size (25 x 16 cm) is classified second class because of its toughness. In both categories the number of spines is low they are soft thus not inflicting too much harm to the grower during management. Another important feature of the leaf is the high content of resins which give the leaf a viscous consistency when boiled. To remove a significant proportion of resin most people use traditional recipes such as adding the skin of a mexican tomato or sodium bicarbonate when boiling. Once boiled, the water is strained and the nopal-vegetable covered with a wet cloth. In frying or broiling, this process is not carried out.

Nopal as a food is a high profile vegetable providing a wide variety of meals unique to the Mexican Basin based on prehispanic precedences (Sanders 1976). Main features of the traditional cuisine is that all the ingredients are boiled then consumed cold. The food may be combined with up to 35 ingredients (80% of cases) dominated by vegetables and spices; other products include local fish, shrimp and insects (20%). Most of the produce is native of the region (70%) or from neighboring regions with similar weather conditions.

Table 5: Proximate composition of silage ingredients (DM basis)
Ingredient	Percentage of:
	DM	CP	EE	CF	NFE	Ash
Nopal-veg	7.5	7.3	2.0	10.8	65.8	14.1
Maize stover	90.6	6.4	0.9	34.8	48.6	9.3
Molasses	63.4	3.0	0.9	-	83.0	13.1

DM Dry matter, CP N*6.25, EE Ether extract, CF crude fibre, NFE nitrogen-free extract

Nopal-vegetable terraces are highly productive as the initial harvest after planting is within the first six months. The age range for the productive plant is five to twenty years. However, a significant proportion of producers reported a mean value of ten years whereas 27% stated they had plots over 15 years of age. Mean yield of a nopal harvest, measured in terms of the number of leaves for the regional mean size of the plots was of the order of 750, 1750 and 5500 units during the low, medium and high seasons of the year, respectively. Accordingly, most productive harvesting is in the winter season during the months of December to January (44%) in contrast with relatively poor harvests during the months of March, April and May (dry season, 11%). For the rest of the year, productivity appeared to be average (66%). Bulk selling of nopal is done in several traditional ways: bunches, dozens, hundreds, a large basket and a large sack. Though producers sell directly to consumers in the market of Milpa, middle men usually buy in quantity which is then taken to Mexico City (the Metropolitan Food Supply; the largest wholesale/retail market in Latin America), to distribute to regional markets in Mexico and for export.

The proximate composition of the material used in the nopal-vegetable silages is presented in Table 5. The DM content of the nopal-vegetable is low showing it to be a very succulent material.

Table 6: Mean values of different silage parameters
	Levels of:
	Molasses			Maize stover
	0	7	15	30	35	Sign.
pH	3.91	3.94	4.02	3.94	3.98
N*6.25#	6.93	6.25	5.52	6.29	6.18	.05
NH3##	8.81	9.02	6.17	7.32	8.80

# % of DM
## % of DM as total calculated nitrogen in each silage.

The values for pH, N x 6.25, and NH₃ are presented in Table 6. As expected, there was no interaction between the levels of maize stover and molasses. With the increase in molasses content in the silage there was a parallel increase in the values for pH and an inverse trend in crude protein and ammonia contents. Higher levels of molasses tended to decrease crude protein content because of the replacing effect of the nopal-vegetable. The addition of the two levels of maize stover increasesd ammonia concentrations similar to the response reported for pH.

Nopal production in the terrace areas is one of the best examples of a crop with a clear adaptation to the environmental conditions of the region. Production is based on the low use of technology which nonetheless provides sufficient economic and social benefits for the local population. It is therefore proposed as an example with high potential of a sustainable model of agriculture in the region. Furthermore, nopal production is one of the few crops that acts as a natural barrier capable of halting the growing urban development of Mexico City, a fact recognised by governmental authorities. Relevant aspects to discuss are those related to the spatial distribution of the crop in relation to the town, combined with the growing demand and the traditional management of nopal, including the intensive use of manure, seasonal patterns of productivity and the potential of the plant to be included as a food to support ruminant production.

Historically, following the fall of Mexico to the Spanish, the conquerors shared the land and supported the development of nopal production because of the natural association of the plant with the endemic insect, cochineal (genus heminoptero), used by local inhabitants to produce natural dyes (Romero 1990). Their inten tion was to dominate the international market in dyes. After the initial colonising period, and the attempted market domination which failed, nopal growing remained without change and possessed additional status beyond the production of leaves for food. However, at the beginning of the 1980's, nopal-vegetable production was favoured with a second boom when the leaves of the agave plant were introduced to the oriental markets of Japan. As a result, the plant now commands a high demand within the country. This raised the national price of nopal, stimulating producers to increase productivity. Moreover, the market was strengthened when literature emerged suggesting that resins present in the leaves have an important hypoglicemic effect. A second associated factor was that related to the high content of natural fibres in the leaves, said to have an effect in preventing colon cancer and/or as a bulk meal to combat obesity (Alarcon et al 1993; Pimienta 1993). The high demand for nopal increased production in the region more than two-fold from a reported value of 2,253 ha in 1984 to 7000 ha to date (COCODER 1985).

The recent increase in productivity was managed by altering the cropping pattern from maize production. This phenomenon also changed the operation of home gardens in the area (Rivera et al 1993), a situation which is unusual in this country as maize is grown as a cultural preference, introducing the nopal-vegetable to the home gardens of the households (SEP 1984). The spatial distribution of nopal within towns and in the surrounding countryside has several reasons. Though the growing areas are managed with a low use of technology, the work that farmers do on the terraces is labour intensive with most chores concentrating at harvest time and the elimination of weeds, thus the locations of nopal growing are usually close to the farmers' homes (Rivera et al 1994). A second reason is to be able to watch over the site as stealing and selling of the stolen leaves is relatively easy. A third factor may be understood in terms of the anthropological concept referring to the appropriation by people of space giving them psychological security (Correa 1993). Finally as the quantity of the nopal taken at any one harvest is smaller than that of the maize harvest, the location of the crop near the home and the town facilitates the need for frequent transport of the crop to market.

According to evidence revealed in the present study there are many facts indicating that producers have continued to use traditional technology to maintain nopal production. In this respect, important processes are involved in the intensive use of manure and the preferred system of planting and managing the agave. The use of manure on the terraces has a duel function; by offering a source of nitrogen and organic matter to the soil this increases temperature at the base of the plant during the winter season. Most of the manure utilised comes from dairy herds managed in peri-urban conditions of Mexico City (Cortes et al 1993). As the presence of manure from these production units is one of the main complaints of people living near the stables, the employment of it in nopal production assists in reducing the problem. Generally, the manure is obtained in fresh form without any cost, the only expence the producer has to outlay is transportation. It is clear the two systems of production have found a way to use waste material to obtain a mutual benefit and, as such, provides clear evidence that the system is sustainable.

A second feature supporting the use of traditional systems refers to the preferences of most of the producers in the region of Xochimilco to maintain production in the customary manner. According to the survey data, 95% of producers preferred the traditional system over a semi-technified approach. Variation between both systems is attributed to plant density per m² and the technique of preparing ridges and spaces between rows. The result of such differences is a higher production of nopal in the semi-technified system which in theory should be the best, a fact that in reality is not necessarily so. The explanation of this phenomenon refers to the seasonal fluctuation in yield of the nopal-vegetable harvest which is in three categories: low, medium and high; and the prices reached in the market with an inverse trend: high, medium and low. In this situation it would appear that most producers prefer to deal with the traditional system to produce less amounts of leaves all year round.

In relation to seasonal variance of production and the actual price nopal achieves in the market, some producers have suggested the idea of feeding ruminants with the surplus production. According to producers' trials, and later in scientific research (Sanchez and Castañeda 1991), the mean number of plants per m² could be increased from 6-8 in the traditional system and up to 16-20 plants/m² in intensive forms of produc tion. This change could increase total yields from the actual figure of 90,000 tonnes to an estimated 200,000 tonnes per year. The decision to offer silage instead of using the fresh nopal for feeding ruminants was based on three main considerations. The seasonal pattern of production only permits fresh nopal to feed the animals during the dry season (March to May). Our own observations with dairy cattle producers from the same zone indicated that they refused to use the nopal-vegetable to feed their animals because of the concequences of diarrhoea and tainted flavour of the milk. Additional factors were the high content of water in the leaves (up to 90%) and the low protein content (Pimienta 1993). Our results demonstrate that the inclusion of molasses and maize stover increased the DM content without any change affecting animal behaviour, as has previously been mentioned in the literature (Huitron et al 1981). In this respect, we are now examining the inclusion of local vegetables which may provide both improved protein and DM content of the silage. The results we have obtained so far indicate a good response. Further steps will involve feeding trials with dairy cattle producers from the zone to test animal production responses.

The results reported show clearly that nopal production in the terrace area of Xochimilco has a high potential as a sustainable model of agriculture. In this respect, nopal cultivation has developed in such a way, adapting to the evironmental conditions of the region with high productivity and involving the work of all members of the family. Although some growers use chemical inputs such as herbicides, insecticides and inorganic fertilizers that may affect the sustainability process, it is clear that the low or zero use of these inputs should be the first step towards maintaining healthy agriculture in the terrace areas. We are now trying to promote with producers the possibility of evaluating a project focusing on the sustainable use of nopal plants of diferent ages, so that they (and we) will have a clear insight to the new proposal. The results of this study will be the subject of future papers.

The authors wish to thank the producers of the nopal-vegetable from Milpa Alta and particularly Mr Silverio Rojas from San Pedro Atocpan and Mr Benito Muñoz for the nopal given for silage making. Ackowledgement is made to the authorities of the Universidad Autónoma Metropolitana (Autonomus Metropolitan University) for facilities given to do the research; and to Hector Martínez G., a Rural Anthropological MSc student, for suggestions to improve the discussion.

(Received 1 May 1996)