Cultivating Diversity: Field Scattering as Agricultural Risk Management in Cuyo Cuyo, Department of Puno, Peru

By Carol Goland, 1993.

Chapter 5 - Land and Land Tenure in Cuyo Cuyo

(page 104)

with papa milli and papa panq'o fields are identical in sequence, if not calendrical timing, to the potato fields of the estancia described above.

Ura Ayllu is thus distinct from Puna Ayllu in its ability to plant early potatoes, and to double-crop some of its potato fields. A more important distinction is that Ura Aylleños have lands in the anexo of Aripo which are used for growing maize. Aripo is located about 12 km downvalley from the parent community of Ura Ayllu. The nuclear community of Aripo lies at approximately 3100 m elevation, some 400 m above the river course. Aripo has its own estancia manda. This manda follows a six year rotation, with the first year dedicated to potatoes, the second to oca, and four years of fallow. Ura Aylleños may plant in the estancia manda of Aripo and, in fact, in the first year of the study, many families augmented their potato land by using Aripo's papa manda. For Ura Aylleños, however, the most important lands of Aripo are not manda, they are the maizales, lower-elevation lands on which maize can be grown (Figure 5.1).

Both within Aripo, and between Ura Ayllu and Aripo (along the valley bottom), Ura Aylleños intensively crop maize. These lands are not manda, as they follow no set crop rotation and fallow sequence. Instead, fields may be planted for as long as soil fertility and production warrant. Fallow may last for as few as two or as many as ten years. The first crop to be planted once a field comes out of fallow is papa panq'o, the late-season planting of early maturing potatoes. Once these are harvested, maize is planted, perhaps along with other crops, such as racacha and yacon (Arraccacia xanthorrhiza and Polymuria sonchifolia, both root crops), habas, and poroto (Phaseolus vulgaris, the common bean). Maize and these other crops will then be grown in each successive year until the owner of the plot determines that a fallowing is needed.

For the fields which are planted in corn each year, field preparation prior to planting consisted mainly of cutting remaining stubble and burning it. Planting, in late September or early October can be accomplished by a single person using the rawk'ana (hoe) or, for large fields situated on flat pampa, oxen may be used. Three "cultivos" are common for maize fields: hallmay, llank'ay, and masiy. Many families have a maize ankacha in March or April.³ The major harvest occurs in June, after the ears dry on the stalks.

Aripo is located at the boundary between higher-elevation tuber production (like that of Ura Ayllu's estancia manda) and more temperate zones, suitable for the planting of maize, habas, poroto, yacon, and racacha. I follow Camino et al. (1981) in describing a small "transitional zone" between the two. In this transitional area, fields are variably used. They cannot be intensively cultivated like the maizales, but in protected spots they can be used for planting maize, provided fallowing is frequent, relative to the maizales found at lower elevations. Like the fields of the Aripo estancia manda, they may be used for planting tubers. Sometimes a first year of potato cultivation (frequently double-cropped) is followed by planting maize along with its associated crops. Or, beginning with the first year of cultivation, potatoes, maize, and the other temperate crops may be intercropped, and in following years polycropping continues, less potatoes.

(page 105)

At the lowest extreme of the altitudinal gradient, some Ura Ayllu families hold lands in the upper jungle (ceja de la selva) near San Juan del Oro (1500 m - 2000 m in elevation). On these lands coffee, coca, and fruits are grown. The latter are generally for family consumption, while coffee and coca are cash crops. Ownership and production of these lands adds complexity to the work schedules of Ura Ayllu families, and it is probably no coincidence that the families most active with production in the jungle are those which are large and thus have able workers who can remain at home while the father and perhaps one child travel to the jungle to tend the lands there.

SUMMARY - THE PRODUCTION ZONES OF PUNA AYLLU AND URA AYLLU

Despite their proximity to one another, agricultural production zones in Puna Ayllu and Ura Ayllu are distinctive. They are most alike with respect to practices in the adjacent estancia mandas of each community. The major differences between the two communities in the estancia are slightly different usages for the third cropping year (more habas in Puna Ayllu; more mixed tuber fields in Ura Ayllu), and in their use of manures (more camelid dung in Puna Ayllu; more sheep dung in Ura Ayllu, applied using wanunas).

Away from the estancia, however, the differences become greater. Puna Ayllu contains yet another complete manda system, the Awi Awi manda. Ura Ayllu has no other manda, although individuals do utilize lands in the mandas of Ñacoreque and Aripo. More significantly, Ura Ayllu households hold non-manda lands in various areas downvalley from the community itself. These provide important cropping opportunities absent both in Puna Ayllu and in the estancia manda of Ura Ayllu itself. At these lower elevations, Ura Aylleños grow maize and other more temperate crops, and they are able to produce a double-crop of potatoes using varieties which mature rapidly. These factors add significant complexity to the agricultural cycle of Ura Ayllu.

In the discussion that follows, I distinguish the different production zones used in the two study communities, as described above. For Puna Ayllu, the distinctions are fairly straightforward: there are the lands of the estancia manda, and the lands of Awi Awi manda. Field which cannot be placed in one of these categories are canchones (enclosed gardens, usually adjoining the house), and several fields in Warini, where bitter potatoes are grown exclusively. These latter two types comprise a small fraction of the total fields documented in this study (n = 4 out of a total of 290 fields in Puna Ayllu).

In Ura Ayllu, the situation is more complex. Like Puna Ayllu, there are the fields of the estancia manda. In addition to these are fields in Aripo and the fields lying between the community of Ura Ayllu and its anexo Aripo. The discussion of Ura Ayllu agriculture will distinguish six types of fields or production zones (from lowest to highest in elevation): maizales (maize and associated crops), transitional (potatoes, maize, and associations), Aripo manda (potato and oca), "Ñacoreque" (potato), "Paqhchani" (potato and oca) and "Ura Ayllu estancia" (papa manda, hatun tarpuy manda, cuti oca manda, and habas manda) (Figure 5.1).

LAND AND LAND TENURE IN CUYO CUYO

Having described the kinds of land which exist in Cuyo Cuyo, we can now discuss how landholdings are distributed in the aggregate sample as well as among individual families. The data used for this discussion are drawn from the field visits and measurements during each of the two study years. In the first year (the 1985-86 agricultural season) 327 fields were in

(page 106)

production for the collective 20 families (ten in each of the communities). In 1986-87, 348 fields were in use.⁴ These numbers represent 488 unique field locations, 227 in Puna Ayllu and 261 in Ura Ayllu. Of the total 488 fields, detailed tenancy data is missing for 20 (5%). In the discussion that follows, when percentages are cited, they are calculated with respect to an n of 468 (total fields with known tenancy).

Most of the fields worked by Cuyo Cuyo families are owned by them. Of all fields documented in this study, only 18% were not owned by the families who cultivated them (Figure 5.2 and Tables 5.1 and 5.2). Fields which were not owned were usually rented. Less commonly, these were fields that were considered to be an "eventual inheritance" from parents. While parents are alive their children may use their fields, but final transfer of these lands will not be made until the parents die. Fully 82% of all fields were owned by those who worked them.⁵ Most of this land was acquired through inheritance (69.7%); only a small portion of it was purchased (10.7% for the aggregate sample). A very few (1.5%) fields were acquired through means other than inheritance or purchase (usually when a man was ceded land by the community when new lands were appropriated).

Clearly, inheritance accounts for most of land acquisition (69.5%). Men appear to inherit more land than their wives (40% vs. 29% of all fields used by families). Moreover, both men and women appear to receive greater numbers of fields from their fathers than from their mothers (29.1% vs. 9.2% for men; 19.8% vs. 10.4% for women, as percentage of all inherited land). Both may inherit land from other relatives (13.3% for men, 0.9% for women), frequently aunts, uncles, and grandparents (Figure 5.3).

Many Cuyo Cuyeños state that inheritance practices follow strict rules of partible inheritance, that is, that each child receives an equal portion of each of the parents' fields. One man enacted such a division for me by stretching out a rope representing the length of one of his fields and then dividing the rope into seven segments of equal length, representing each of his children, and indicating that this practice would be followed for each field. It seems likely, though, that this practice is not always or even often the case, for, as small as some field parcels are, they would be miniscule if such divisions were made in each generation. An explanation closer to reality (as opposed to the conceptual principle of equity), may have been offered by another informant, who stated that size of fields determined whether or not they would be subdivided among children. If small, plots would be inherited whole, while larger plots were divided between children.

Informants also state that sons and daughters are treated equally in inheritance, although the greater proportion of fields coming into a marriage from the husband suggests that this may not always be the case. For all inherited fields, 58% are brought into the marriage by the husband, while 42% were brought in by the wife (Figure 5.4). While both men and women receive fields from both of their parents, both receive more from their fathers than their mothers. This suggests, as does the preceding discussion, that, at least for the two generations involved here, men possess more land. For all men's fields for which the source

(page 111)

of the inheritance is known (168.5 fields), 56.5% were inherited from their fathers, while 18% were from their mothers and 26% came from others (Figure 5.5). For women, where the source of their inheritance is known (101.5 fields), 63.5% of fields were received from their fathers, 33.5% from their mothers, and 3% came from others (Figure 5.6). Viewed another way, fathers are the source of a majority of inherited land, 59%, while mothers are the source of 24% of it, and others account for 17%.

This may suggest, then, that contrary to the claims of Cuyo Cuyeños, males may receive some favored treatment in the inheritance of land (given that they bring almost 58% of the inherited fields into a marriage as opposed to 42% for women). From the perspective of the progenitor, fathers pass about 60% of their fields to their sons, vs 40% to their daughters (Figure 5.7). On the other hand, these data also suggest that for whatever land women do hold, daughters are the favored recipients, since mothers pass 47% of their fields to sons and 53% to daughters.

Tables 5.1 and 5.2 (see also Figures 5.8 and 5.9) summarize tenancy data for each informant household in Ura Ayllu and Puna Ayllu, respectively. Most families utilize lands from all potential sources: rental, inheritance, and purchase, although the proportions of the mix vary, sometimes dramatically. Family A may be "typical" in that (1) all land sources are represented, (2) the predominant form of land acquisition is through inheritance, and (3) the husband has brought a disproportionate percentage of the inherited land into the marriage. Note that a "land-rich" family such as G rents no land, despite the fact that they are one of the largest families in the study (with 7 children still at home). These figures, of course, only represent the number of fields and not their sizes. In fact, of all families, Family G planted by far the greatest amount of land in each of the study years (nearly 21,000 square meters in 1985-86; more than 13,500 in 1986-87). Families H and I illustrate the consequences of being foreign to the community. The husband in Family H is from outside of the community. Thus, he brought no lands to the marriage and this family depends to a large extent on rented lands. The same problem exists for Family I, although its foreign origin is one generation back. That the husband in Family I brought no land to the marriage is explained by the fact that both of his parents were born outside Ura Ayllu, and were brought there as colonos to work the lands of the hacienda.

In Puna Ayllu (Figure 5.9) families exhibit much the same sort of pattern and variety. Most families utilize lands from various sources, and men bring more fields to the marriage through inheritance than do women. The family with the greatest amount of land in production (Family O: over 9,000 square meters in 1985-86 and more than 8,000 in 1986-87) is also one of the few that does not resort to land rental. Although Family P's land totals are not as striking, they are an elderly couple with limited production needs who actually rent out a portion of their lands. Family Q is the only household with one spouse foreign to the community. This is seen clearly by the fact that they must rent a high proportion of their land to meet consumption needs. The large amount of unowned land used by Family M is "eventual inheritance," considered on loan from parents of both spouses.

It appears that in Puna Ayllu a smaller proportion of families use land which they do not own. These generally are young families waiting for eventual inheritances (like Family M), or families where one spouse (or parents of the spouse) was foreign to Puna Ayllu (e.g., Family Q). Ura Aylleños rent lands in neighboring communities (e.g., Ñacoreque), a practice that appears to be absent in Puna Ayllu. Finally, it should be noted that most of the "other"

(page 117)

category in Ura Ayllu are lands ceded by the community from liwa.⁶ In Puna Ayllu, there have been no liwas, at least in recent times, and the only "other" source of land was from the deceased spouse of a first marriage.

THE FIELDS OF PUNA AYLLU

In 1985-86, the ten study families in Puna Ayllu planted a total of 146 fields (Table 5.3); the following year 156 fields were planted by the same families. The minimum and maximum number of fields among all families was, in 1985-86, 9 and 20, respectively, and in the following year, 11 and 25 (Figure 5.10). For the community as a whole, 32,238 m² were planted in 1985-86, and in the following year, 38,615 m² were in production. By family, the maximum amount of land planted in 1985-86 was 9,463 m² and 8,225 m² in 1986-87 (in both years, Family O), while the minimum amount of total land planted was, in 1985-86: 1,265 m² and in 1986-87: 1,945 m² (Family K in both years) (Figure 5.11). The average size of fields increased from 216.5 m2 in the first year to 258 m² in 1986-87.

The average elevation for all fields was 3707 m in 1985-86 (with a range of 3470 - 4100) and 3747 m in the second year (range: 3470 - 3985) (Figure 5.12). The increase in average elevation is undoubtedly due to the progression of the mandas, since in 1986-87 both the estancia and Awi Awi papa mandas were at particularly high elevations. Looking at these data according to field location, the average elevation of estancia manda fields was 3655 m in 1985-86 (n = 88⁷ ), and 3700 m in 1986-87 (n = 94). For Awi Awi manda fields in 1985-86 the average field elevation was 3792 m (n = 56), and in 1986-87 the average rose to 3818 m (n = 54).

In 1985-86, of the 146 fields planted by Puna Aylleños, 65.5⁸ (53.3%) were brought into the marriage by the husband, 34.5 (28.0%) were inherited by the wife, 21 (17.1%) were purchased, and 1 (0.8%) was acquired through other means. A total of 23 fields were rented or "borrowed" (Table 5.3 and Figure 5.13). In the following year, for the 151 fields for which there are tenancy data (out of the total of 156), 55.5 (44.4%) were brought into the marriage by the husband, 46.5 (37.2%) by the wife, 19 (15.1%) were purchased, and 4 (3.2%) were acquired in other ways. The remaining 26 fields were rented or borrowed.

THE FIELDS OF URA AYLLU

In both years of the study, Ura Aylleños planted a greater number of fields and a greater total field area than their Puna Ayllu neighbors (Table 5.3). In 1985-86, the ten study families in Ura Ayllu planted a total of 181 fields; in the following year, the total number of fields was 192. The minimum and maximum number of fields among study families was, in 1985-86, 11 and 24, and in 1986-87, 10 and 26 (Figure 5.10). For the community as a whole,

(page 124)

49,516 m² were planted in 1985-86, and in the following year, a total of 46,060 m² were planted.⁹ The family with the maximum amount of land in production (Family G) planted an unusually high total of 20,707 m² in 1985-86. The same family also planted the greatest area in the following year, but its total was a more modest 12,214 m² (Figure 5.11). The average elevation for all fields was 3385 m in 1985-86 (with a range of 2705 - 3665) and 3406 m in 1986-87 (range: 2750 - 3820).¹⁰ (Figure 5.12). Average field area fell slightly from 263 m² in the first year to 240 m² in 1986-87.

In 1985-86, of the 179 (out of 181) fields for which land tenancy data are available, 80 (49.4%) were inherited by the husband, and 55 (34.0%) by the wife. There were 21 (13.0%) purchased fields, and 4 (2.4%) fields acquired through other means (primarily through distribution of liwa by the community; see footnote 6). Of all fields, 17 were rented or borrowed (Table 5.3 and Figure 5.13).¹¹ In 1986-87, tenancy data is available for 184 of the total 192 fields. Of these, 70 (47.3%) were inherited by the husband; 55 (37.2%) were inherited by the wife, 20 (13.5%) were purchased, and 2 (1.4%) were acquired through other means. The total number of rented fields was 36.¹²

THE DISTRIBUTION OF LANDS

As described above, the manda system dictates to a large degree which crops will be sown in a given field in a given year. For each family, this means that both the amount of land and the crop mix are largely determined by how the family's lands are distributed across the manda sectors. This distribution will be a decisive factor in determining the quantities and proportions of locally-grown foods available for consumption over the next year. In addition, extreme inequities in holdings over various sectors may create serious shortages in seed for the successive planting, which may in turn have consequences for consumption. If a family had a very small landholding in the papa manda in one year, and a large holding available in the manda sector destined for potato the next year, a relatively greater proportion of the potato harvested in the first year would have to be set aside for seed, further diminishing an already small stock available for consumption. For example imagine that a family's holding of 300 square meters produced 255 kg of potatoes (given a yield of 8500 kg/ha), and imagine that in the following year the family had 800 square meters available to plant potatoes. The family would need 184 kg of potato seed (given the average seeding rate of 2300 kg/ha for potatoes in Cuyo Cuyo). In consequence, only a little more than one-fourth of their potatoes would be available for consumption. In contrast, if the same quantity of land (300 square meters) were planted in potato, nearly three-fourths of the previous harvest could be consumed.

(page 125)

Given these considerations, we might expect that Cuyo Cuyeños would attempt to hold similar land areas in each of the manda sectors. I examine this issue in the following sections.

PUNA AYLLU

Table 5.4 (see also Figure 5.14) presents aggregate data for the distribution of landholdings in Puna Ayllu's estancia manda over the two study years. Note first that the amount of land in 1985 in the papa manda (Year 1) is virtually identical to that of the oca manda (Year 2) for 1986. This is what we would expect, since they are the same set of fields. This observation holds true for a comparison of Year 2 from 1985 and Year 3 in the rotation from 1986, although in both cases, there is a slight drop in the aggregate total. The amount of land in Year 3 of the rotation in 1985 and Year 4 in 1986 show but a slight increase (about 100 m²).¹³ Thus, there are no surprises when the land distribution is examined this way, although it confirms a high degree of accuracy in the information, since any large inexplicable changes from year-to-year would suggest that a significant number of fields were omitted in one of the years.

A more important issue is the amount of land available for each crop in each year. In the estancia manda (referring again to Table 5.4 and Figure 5.14), it is apparent that the amount of papa manda land more than doubled between the two years, for the aggregate sample. At the level of the individual family (Table 5.5 and Figure 5.15) the amount of change, at least for most families, is not so great. The amount of land used changes from as little as -3% (Family L) to as much as +1127% (Family T). The average change for families in the amount of potato land is +340% (66% leaving aside the problematic Family T). As suggested above there is a heightened potential for shortage of seed in the second of the two years under these circumstances.

The potato lands of the estancia are, however, not the only papa manda of Puna Ayllu. Additional seed is available from the Awi Awi manda. As is apparent in Figure 5.16 (Puna Ayllu Awi Awi lands), the aggregate amount of potato land in Awi Awi scarcely varies between the two years, and per family, the average percent increase is less (about 50% whether T is included or not) (Table 5.6 and Figure 5.17). If we view the amount of total potato land per family (that is, estancia and Awi Awi combined), the average change from Year 1 to Year 2 is 76% (42% excluding Family T). To some degree, then, the change between years in one manda is dampened by the other (see Figure 5.18). During the study years, a large increase in potato land in the estancia manda was offset, at least in part, by smaller changes in the Awi Awi manda. In other years, the potential problem of seed shortage may be buffered if an increase in potato land in one manda is accompanied by reduction in the other. The problem will be exacerbated if both mandas undergo large increases simultaneously.

All land which falls in the manda in the first year of the rotation is dedicated to potato. Likewise, all land in the second year of the rotation is planted in oca. (Compare Table 5.5 and

(page 135)

5.7; Figures 5.19 and 5.20). As noted above, the third year of the rotation is used in various ways. Infrequently (although apparently the norm in previous times), this year may be dedicated to a repeat sowing of oca. Or, the fields of this year of the rotation may be polycropped with habas and the minor tubers (oca, isaño, illaco). Finally, these fields may be planted exclusively with habas. In the latter case, the fourth year becomes a true cuti habas manda (repeat habas). This divergence is apparent in comparison of Figures 5.19 and 5.20. The first figure shows the land distribution, by rotation year, for families in 1985. The latter indicates the actual crop planted. For Years 1 and 2 in Figure 5.19, and crops papa and oca in Figure 5.20, the field areas are nearly identical in the case of each individual family. On the other hand, the areas devoted to habas and other tubers show clear elasticity. For example, Family L uses a portion of its third year rotation for other tubers, but obviously a small portion of third year fields are cropped in habas alone. In contrast, families P and S forego cropping any tubers in the third year of the rotation, and instead dedicate all of their third and fourth year fields to habas.

The same pattern of divergence is apparent in 1986 (Tables 5.5 and 5.8; Figures 5.21 and 5.22). The amount of land contained in the first year manda is the same quantity of land used for planting potatoes. All land in the second year of the rotation is used for oca. Some third year land may also be used for oca (i.e., "cuti oca" - compare Family M: more land area is cropped in oca than just the second year fields). Some families (e.g., R) use the third year for minor tubers, but in almost all cases, families plant habas, and habas and tuber intercrops, for the third year of the rotation instead of repeating tuber crops in the estancia.

How fields in the third year of the rotation are used is an aspect of estancia agriculture that is significantly different between the two communities. In Ura Ayllu, comuneros plant the third year of the rotation in cuti oca, that is, a repeat oca crop, and only one year--the fourth and final--is used for habas. In Puna Ayllu many fields in the third year are cropped in habas. The cropping choices for the third year of the rotation are one of the keys to viewing all of the lands of each of the communities, and to understanding the inter-relations between rotation years.

The preceding figures seem to indicate that if any crop is favored, it must be habas. When the fields of Awi Awi are included however, we see that this is clearly not the case. Potato and oca are the predominant crops in Puna Ayllu. For most families, only about 25% or less of all land is planted with habas. (Tables 5.9 and 5.10 and Figures 5.23 and 5.24 for 1985; Tables 5.11 and 5.12 and Figures 5.25 and 5.26 for 1986). This suggests that precisely because Puna Aylleños have additional lands for cropping ocas (outside of the estancia) they are able to forego the repeat cropping of it. The opportunity to grow oca in the Awi Awi manda means that the habas crop can expand into the third year of the estancia rotation without creating a deficit of oca. I shall argue later that the situation is the reverse in Ura Ayllu.

URA AYLLU

URA AYLLU ESTANCIA MANDA

Table 5.13 and Figure 5.27 present data on the aggregate distribution of Ura Ayllu estancia manda lands across years of the rotation cycle, comparing each of the two years of the study. Comparison of Year 1 for 1985 with Year 2 for 1986 (and so on) again provides a rough evaluation of the accuracy of the data since the fields in the first year of the rotation in 1985 become the fields in the second year of the rotation in 1986. Their totals should be identical, or nearly so.

(page 148)

The changes are of small magnitude: a drop of some 200 sq m from Year 1 of 1985 to Year 2 of 1986; a larger drop (almost 600 sq m) from Year 2 of 1985 to Year 3 of 1986; and a slight increase (less than 200 sq m) from Year 3 of 1985 to Year 4 of 1985. These trends are disaggregated in Figures 5.28, 5.29, and 5.30 (see also Table 5.14). Referring to these, it is possible in most cases to explain household gains or losses of land. For example in Figure 5.28, the gain in Family C's lands is, I suspect, a true data omission. Two fields which were declared in the 1986-87 field visits as in the second year of the rotation had not been reported the previous year, when they presumably were planted in potato. In the second year of the study, Family D failed to plant one field which should have been planted, following the rotation, as part of the hatun tarpuy oca manda. Family I, a land-poor household, was able to rent a potato field in the estancia from the community in 1985 for that year only. In the subsequent year, they had no land in the hatun tarpuy oca manda. All other families planted the same area in the second year of the rotation of 1986 as they had in the first year of the rotation in 1985.

In Figure 5.29, the same comparison is made between the land in the second year of the rotation (hatun tarpuy) in 1985 and in the third year of rotation (cuti oca manda) in 1986. Again, most families realize no gain or loss. Most changes are slight, with one exception. The fields of Family J show the largest change. Two of the fields planted in the hatun tarpuy oca manda in 1985 were intentionally not planted in the following year. The reason given for this, explained the family, was that the fields were located in an area of poor quality, rocky soil. In addition, they pointed out that other families with fields in the same area also left them unplanted. The absence of crops in the surrounding area meant that animals would be grazed there, exposing any standing crop to further reductions in yield, beyond those created by the poor quality soil. Family J's drop would be more dramatic, but for the fact that in 1986 they rented one field for the cuti oca manda. Presumably, this was to offset the reduction in field area created by the two problematic fields. The modest increase for Family C is due to a suspected omission in field declarations in 1985. The small decrease for Family E is from a field which had been rented in 1985 but not in 1986. The same factor explains the slight drop in Family H's land area.

In Figure 5.30, field area in the third year of the rotation in 1985 is compared to those in the final year of the rotation (habas manda) in 1986. Again, it is possible to explain the differences between years on a case-by-case basis, reducing the possibility that there are problems of omission with the data. The land area for Family B's fourth year fields in 1986 increased because they purchased a new field. Family C had five cuti oca fields in 1985 but only four habas manda fields in 1986. One field was passed directly to fallow without a fourth year of cropping, for unknown reasons. Family F apparently failed to declare two fields in the third year of the rotation in 1985. When they came to light in 1986, the family was emphatic that they had also been planted in 1985. Other than the fact that they were planted, there are no other data for them from the first year of the study. The land planted by Family H increased from 1985 to 1986. This is due to a new rental. The reason for the reduction in I is less clear. Family J failed to plant one field which was in production in 1985.

Returning to Figure 5.27, we see that, like the pattern described for Puna Ayllu above, the aggregate amount of land in the papa manda about doubles between the two years. This may reflect real differences in the size of the manda sectors, or another explanation may be responsible. Here I have presented the total field area in use by the sample families, ignoring land tenure. In fact, most of the aggregate increase in potato landholding can be explained by a dramatic increase for one individual household, Family H. The amount of papa manda land owned by this family showed almost no change between the two years. On the other hand, the amount of rented papa manda land increased over 1500 sq meters. There is probably little real change in the size of the papa manda between the two years. The drastic reduction in land in

(page 153)

the second year of the rotation actually occurs, while the generally equivalent land areas in Years 3 and 4 are as would be expected.

Tables 5.14 and 5.15 (see also Figures 5.31 and 5.32) present, for each family, the field area in each year of the estancia rotation (Table 5.14 and Figure 5.31) versus the field area for each crop type (Table 5.15 and Figure 5.32) in 1985. The rotation in Ura Ayllu's estancia manda is: papa manda, hatun tarpuy oca manda, cuti oca manda, and habas manda. By crop and crop mix, we would expect the following in each of these years: year one, potatoes only; year two, oca only; year three, oca or oca and other tubers, perhaps with some habas; and year four, habas alone or habas and minor tubers. Comparing the two figures, it is apparent that year one and papa are equivalent for each family. With respect to the remaining years of the rotation, Family H perhaps comes closest to the idealized "potato - oca - mixed tuber - habas" rotation. Few families plant all of their second year fields exclusively in oca (E, F, and H are the exceptions). This is related to the general dissatisfaction Ura Aylleños have with their oca yields. Almost all plant habas in at least the same area as that of the fourth year of the rotation, and many plant additional habas, alone or intercropped with the minor tubers, in the third year of the rotation as well. The residual "other" usage is a planting of oats and barley in the fourth year of the rotation by a family (D) with especially great forage needs.

Tables 5.14 and 5.16 (Figures 5.33 and 5.34) present the same data, for the second year of the study. The same patterns hold: all first year land is planted in potato (and potato only appears in first year lands). It would appear that in this year more of the cuti oca fields are planted only in oca (perhaps related to the reduction in area seen in 1986 for the hatun tarpuy oca manda - see Figure 5.27 and discussion above). Habas are planted in equivalent, or greater areas than the area covered by fields in the fourth year of the rotation. And for most families, fields with other tubers or habas and tuber mixes account for about the same land area as that held in the third year of the rotation cycle.

OTHER LANDS OF URA AYLLU

There are five other production zones used by Ura Aylleños: Paqhchani, Ñacoreque, Aripo Manda, Transitional Fields, and Maizales. Tables 5.17 and 5.18, and Figure 5.35 show the amount of land used by study families in these locales in each of the study years. The first three field types are important for growing potatoes--especially early-maturing varieties--while the latter two are significant for production of maize.

The Ñacoreque fields are part of the manda of the community of Hatun Ñacoreque, which is located about 6 km from Ura Ayllu. The lands used here by Ura Aylleños are rented, except for Family F (the wife is from Ñacoreque and thus has inherited lands there). Fields in the Ñacoreque manda are sown for one year only, with potatoes, followed by 5 years of fallow. In the Aripo manda, there are two years of cultivation followed by four years of fallow. The cropping sequence is potatoes followed by oca. Here Ura Aylleños own rather than rent lands, but they generally do not use the second cropping year. The Paqhchani fields are part of Ura Ayllu and located quite close to it, however these lands are not incorporated into the estancia manda. They rotate between potato, oca, and fallow, in a cycle which is not synchronous with the estancia fields (a number of informants claimed that there were three years of cropping usage, followed by as many as eight years of fallow).

With reference to Figure 5.35, the Paqhchani, Ñacoreque, and Aripo manda fields each vary considerably in total amounts planted in each of the two years. On the other hand, if we contrast the Aripo manda fields with those of Paqhchani and Ñacoreque combined, the area of land used in the first year versus the second year is similar (in 1985-86 there is heavy