Livestock Research for Rural Development 14 (1) 2002

Characterization of the productive performance in family pig farms located in Ayotzingo, State of Mexico

D Mota, R Ramírez-Necoechea, M Alonso-Spilsbury and A C García-Contreras

Agriculture and Animal Production Department,
Metropolitan Autonomous University-Xochimilco (UAM-X).

Ecodevelopment of Animal Production,
del Hueso 1100. Col. Villa Quietud. Mexico, DF 04960.



The aim of this study was to characterize the productive performance of  family pig farms in the Ayotzingo community, State of Mexico. The data compilation was made in two stages; the first by a technique based on interviews following a previously elaborated survey questionnaire; the second by visits every 15 days during a period of seven months to record and register productive and reproductive indicators.

The average number of piglets born alive and stillbirths per litter were 9.1 and 2, respectively. Both traits were affected by the parity of the sow. The individual weight at birth was 1.3 kg and the number of weaned piglets 7.02. Age and weight at marketing were 210 days and 86 kg. The feeding in the farms was characterized by kitchen wastes (80%) and the remainder (20%) as balanced feed. Feed conversion and daily weight gain were on average 5.5 and 412 g. From the total of weaned sows showing estrus (150), 76% became pregnant (114), and the remainder repeated the cycle. Out of the 36 that repeated estrus, 4 (2.6%) did not receive artificial insemination or direct service because there was no chance to get a boar on time. It is important to mention that 143 sows (95.3%) received natural service and only 7 (4.3%) were artificially inseminated.

In spite of the inadequacy of the feeding systems, husbandry and equipment, the results of the study indicated that backyard pig farmers can obtain positive productive levels of performance.

Key words: Native pigs, family pig farming, growth, reproduction


Rural and peri-urban pig farming is a form of production characterized by a low scale of activity basically of subsistence (Cuarón 1987; Losada et al 1997; Ramirez et al 1998). This type of farming is mostly handled by old men, women, and children, which is why it is known as “family pig farming” (Fickers 1991). Producers that are dedicated to this kind of husbandry usually have from one fattening pig to several sows. The genetic quality of these animals is low, although the rusticity and adaptability to the environment enables them to produce meat with a minimum of “balanced” purchased feeds.  They usually are fed with kitchen wastes, grains such as maize, wheat by-products, alfalfa and frequently scavenge on available grazing (Castillo 1988; Conejo and Mejorada 1990; Losada et al 1995). Pigs in this system have low levels of production, through lack of knowledge on the part of the producers (Ramírez 1997), as well as inadequate marketing opportunities (Fickers 1991; Conejo and Ortega 1995).

Backyard husbandry systems must be considered as a peculiar productive stratum within the context of national pig farming. They are unlikely to disappear and therefore cannot be ignored, and so must be studied in greater depth to know better the levels of productivity and the limitations to achieving higher rates of production (Suárez 1995; Ramírez and Mota  2000).

In intensive pig farms it is common practice to routinely evaluate the system in the pursuit of greater productivity. By contrast, there is a lack of information concerning levels of productivity in family pig farms (Ramírez et al 1999a; Mota et al 2000). Yet this type of pig husbandry has many owners. There are more than 1 million 300 thousand family productive units according to INEGI (1990) and (Ramírez 1997) that contribute over 30% of the national swine inventory.

Performance in a swine enterprise should be guided by specific target levels, with the aim of obtaining higher profits through an adequate business control. When a pig farmer wants to have a wide vision of the farm productivity, it is essential to apply methods that illustrate this through the productive indicator analysis (Ramírez et al 1999b). To achieve this the producer needs a record containing all information necessary to realize such an evaluation. Family pig farms generally do not have records that let them quantify animal production; moreover, they rarely have access to technical support to guide them in this task (Fickers 1991). As a result the family pig farmer often does not know if the production numbers are good or not (Suárez and Barkin 1990).

The objective of the present study was to characterize the productivity and performance of the family pig farms in the Ayotzingo community in the State of Mexico.

Materials and methods

The observations were made from July 2000 through January 2001. Production data in 25 family pig farms located in the Ayotzingo community in the State of Mexico were recorded and analyzed. The type of pig that prevailed in the study was the “native” (an assortment of crosses between Creole and improved breeds) (Photo1). Farms that had only “improved” breeds were not included in the study.

Photo 1: A typical crossbred sow with crossbred progeny

The data compilation was made in two stages; the first by a technique of structured interviews (Ander-Egg 1974) following a previously elaborated questionnaire. The aim was to obtain information to characterize the production system, and included husbandry, inventory, feeding and productivity activities. Farmers were also helped to fill up the individual records of the animals. The second stage consisted of visits every 15 days during a period of 7 months to verify and record the productive and reproductive indicators of the farms where the studies were done.

The following parameters were registered: number of piglets born alive, number of stillbirths, individual and litter weight at birth, and at weaning; number of weaned piglets, pre-weaning mortality; weaning age, percentage of stillborn piglets, feed conversion, and age and weight at market. The methodology used to evaluate and characterize the “runt” pigs was that described by Ramírez and Mota (1999).

Reproductive parameters registered were: interval from weaning to first service, interval from weaning to effective service, lactation period, fertility percentage, abortion percentage, productive life span, productive cycle, and interval between farrowings.

The statistical analysis of the data consisted of central trend measurements. An analysis of variance was performed to evaluate the effect of parity on the number of piglets born alive and dead. The model included parity and error with farms as replicates.

Results and discussion


The average of females per producer was 6, with ranges between 2 and 12 sows. The productive life was relatively short with 84% of the sows farrowing only three times (Table 1).  

Table 1. Sow distribution by parity


Number of sows





















Productive performance

The results of the present study (Tables 2) were compared with other national studies where native or crossbred pigs under different confinement conditions were evaluated, considering the productive performance that characterize family pig farming. We also took into account the minimum range of productive indicators that have been estimated for semi-intensive production system farms in Mexico (Batista 1993).

Table 2. Productive performance of the sows and growth performance of the weaners (mean and standard deviation) (n=190)




Litter performance



Piglets born alive






Litter weight at birth (kg)



Individual weight at birth (kg)



Piglets weaned



Total weight at weaning  (kg)



Individual weight at weaning (kg)



Weaning age (days)



Mortality of piglets during lactation



Uneven litters (runts) (%)



Post weaning performance



Weight gain (g/day)



Feed conversion (kg DM/kg weight gain)



Days to market



Weight at marketing (kg)



The average number of piglets born alive (9.1) was higher than reported by Batista (1993) and Góngora et al (1986). However, it was in the lower range of the findings reported by Trujillo (1998) and similar to those observed by Vázquez et al (1972). There was an average of 2 stillborn piglets per litter representing 19% of total births. Both before-farrowing and intra-farrowing mortalities were included. Mota and Ramírez (1997) reported that the sum of both mortalities is normally in the range of 10 to 12% and should never exceed 12%. The reason for the high percentage of stillbirths in our study could be related to the limited skills of the producer to adequately attend the parturition, combined with inadequate nutrition during pregnancy.

Another feature to consider was the indiscriminate use of oxytocin. We found that 19 out of 25 interviewed producers (76%) used this hormone during farrowing. According to Mota and Ramírez (1998) and Mota et al (2001), oxytocin should only be administered during parturition at specific instances. When used routinely it can produce higher intra-partum stillbirths due to intermittent and violent uterine contractions that break the navel cord before time, provoking asphyxia.

The number of piglets born alive and the stillbirths were affected by the parity of the sow (Figures 1 and 2). The highest number of piglets born alive was for sows in their 3rd, 4th and 5th parity.  The highest number of stillbirths was for sows in their 5th and 6th parities, followed by the 1st parity. This agrees with reports by Mota and Ramírez (1997) based on data from intensive pig farms.


Figure 1
Piglets born alive  (means and standard errors) according to the parity of the sow.


Figure 2.
Stillbirths (means and standard errors) according to parity of the sow.

The mean weights of the litters at birth (11.6 kg) and of individual piglets (1.3 kg) were within the range reported by Trujillo (1998). The number of piglets weaned (7.02) was slightly less than the minimum range reported by Batista (1993) and Trujillo (1998) but similar to results reported by Vázquez (1972) (7.0) and higher than the figure (4.3) quoted by  Góngora et al (1986). This indicator is linked to the pre-weaning mortality, which at 22.7% was higher than accepted norms of 12% during this period. Ramírez et al (1999b) concluded that this parameter must not exceed 20%. The most important causes of pre-weaning mortality were reported to be pneumonia and dehydration by diarrhea.  Factors that could explain the mortality were the lack of an appropriate microclimate for the piglets, such as provision of bedding to let the animals thermo-regulate themselves. Other factors were related to the undernourishment and the poor body reserves of the lactating sow, both of which contribute to low milk yields, and the lack of creep feed for the nursing piglets. All of these problems are typically found in family pig farms in Mexico. The percentage of uneven litters was 42%. This is a common problem for backyard producers.

Considering that creep feed was rarely provided, the mean litter weight of 96.2 kg and the individual weight at weaning of 9.5 kg were within the normal ranges reported for native pigs kept under tropical conditions (Vázquez et al 1972). The time to reach a marketable condition and the live weight achieved are determined by the growth rate. The recorded mean growth rate of 412 g/day is reasonable, taking account of the fact that the animals in this study were crosses with native pig strains and that feeding was based on kitchen wastes. For comparison, the growth rates in intensive pig farms are of the order of 700 g/day while Mexican hairless pigs were reported to gain 140 g/day (de Dios et al 1989; Alonso-Spilsbury et al 2001). The estimated dry matter feed conversion of  5.5 must also be related  to the feeding system, since 80% of the feeds in these farms were kitchen wastes (leftovers) and only 20% were grains and balanced feed.

Reproductive performance

The interval from weaning to estrus (Table 3) was longer than that reported by Leman (1990), Mota et al  (1997) and Trujillo (1998). Under normal conditions sows show estrus within 10 days post weaning. However, when the nursing sow is under-fed, she produces milk at the expense of her body reserves and rapidly loses body condition with consequent delay in initiation of the next reproductive cycle. This, undoubtedly, is one of the critical problems to solve in family pig farms. The solution is better nutrition during pregnancy, but the extra cost this entails must be balanced against the benefits from a reduction in the non productive period and the potential increase in production of piglets per sow per year (Mota et al 1997; Ramírez et al 1999c).

Table 3. Reproductive performance (n = 150)




Weaning to estrus  (days)



Weaning to conception  (days)



Lactation (days)



Number of abortions



Life span (number of parities)



Fertility (%)


62 – 79

Interval between farrowings (days)



 It was observed that of the 150 sows and litters studied, 93% showed poor body condition at weaning. We consider this to be the main cause of the delay to first estrus after weaning. Of the total of weaned sows showing heat (150), 114 became pregnant (76%) and the remaining 24% repeated estrus. From the 36 sows that repeated estrus, 4 (2.6%) did not have the opportunity to conceive as neither artificial insemination nor natural service was available at the needed time. The interval from weaning to conception (43.6 days) was only slightly longer than the interval from weaning to first estrus. This indicates that the problems were not because the sows did not conceive but that they took longer to express estrus. The outcome was an extended interval between farrowings of 203 days, equivalent to 1.8 farrowings per sow per year. Probably this is the most important indicator of reproductive performance, which in the present study could mainly be traced to inadequate nutrition. However, as stated earlier, this is an economic issue in which the cost of additional, or more balanced feed  must be set against potential benefits in terms of piglets produced per sow per year.

Natural service was the preferred method of conception with 95% of the sows in this category. Abortions were not a problem with only 1.3% incidence. A surprising result was that 84% of the studied population were young sows that did not exceed 3 parities and only 10% of the population exceeded 5 parities. This may also reflect effects of poor nutrition as presumably older sows were culled because of poor reproductive performance.


There were obvious deficiencies in the feeding system, the husbandry conditions and equipment in most of the family pig farms taking part in the survey. Nevertheless, levels of productivity were close to national averages.  Reproduction was the weakest link in the production system, almost certainly because of nutritional deficiencies imposed by dependence on kitchen wastes as the main feed.  On the positive side, the use of recycled wastes is an important contribution to sustainable resource utilization, as well as being a low cost system for the producer.


The authors wish to thank the DVM students from UAM-X (R A Gordillo, V Meza, D E Zamorate and M N Flores, for their collaboration. The support of  Francisco Osorno S,  Director of Economic Development, Chalco Municipality, is acknowledged for his efforts to secure the cooperation of the producers, which made this study possible.


Alonso-Spilsbury M, Mayagoitia L, Escobar I, Ramírez N R y Mota D 2001 Foraging behaviour of a Mexican Hairless pig herd kept under agro-forestry conditions. 35th Congr. of the International Society for Applied Ethology. Davis, CA. USA.

Ander-Egg E 1974 Introducción a las técnicas de investigación social, Humanytan Buenos Aires, Argentina, 109 pp.

Batista L 1993 Rumbo parámetros productivos y calidad. Desarrollo Porcícola, 12: 9-15. México.

Castillo J 1988 El cerdo de traspatio. Tecnología Avipecuaria en Latinoamérica, 1: 32-35.

Conejo J J y Mejorada A 1990 Dietas utilizadas en la alimentación del cerdo de traspatio en tres comunidades de Tarímbaro, Michoacán. Memorias del XXV Congreso Nacional AMVEC. México. pp. 244-245.

Conejo J J y Ortega R 1995 Problemas de la porcicultura rural de traspatio en los países en desarrollo. En: Kato L (ed.). La Producción Porcícola en México: Contribución al Desarrollo de una Visión Integral. Univ. Autónoma Metropolitana-Azcapotzalco. Univ. Michoacana San Nicolás de Hidalgo. México. pp. 197-223.

Cuarón I J A 1987 Desarrollo de esquemas tecnológicos para la porcicultura rural familiar. Síntesis Porcina,  6(9): 41-45.

De Dios O, Santos L y Zulueta M 1989 Sistemas de alimentación de cerdos en Tabasco. Síntesis Porcina (Ago.):42-48.

Ficker A M 1991 Pig farming in rural areas of Gabon. Word Review Animal Zootechnie. No. 1, pp. 59-63.

Góngora G S, Richards M y Verdugo R J 1986 Análisis económico y social de la porcicultura rural de traspatio en los municipios de Mérida y Uman del Estado de Yucatán. Técnica Pecuaria México, 50: 115-126.

INEGI 1990 Compendio de información estadística del sector agrícola. México.

Leman A 1990 Cubra las cerdas una vez a los 3 a 5 días del posdestete. International Pigletter, 10(8):19-31.

Losada H, Neale M, Vieyra J, Soriano R, Rivera J, Cortés J and Grande D 1995 The potencial of traditional systems of pig production in the temperate region of Xochimilco. Livestock Research for Rural Development, volume 7, number 1:

Losada H, Pealing R, Cortés J and Vieyra J 1997 The keeping of poultry and pigs in the backyards of the urbanised areas of Iztapalapa (east of Mexico city) as a proposal for sustainable production. Livestock Research for Rural Development, volume 9, number 3:

Mota R D, Ramírez N R y Cisneros P M A 1997 ¿Quiere aumentar su productividad? La clave esta en reducir los días abiertos. Agronegocios en México, (8):15-18.

Mota R D y Ramírez N R 1997 Observaciones clínicas sobre lechones nacidos muertos intra-parto. VII Congreso Latinoamericano de Especialistas en Cerdos. Río Cuarto, Argentina. p. 139.

Mota R D y Ramírez N R 1998 El parto de las cerdas. Nuestro Acontecer Porcino, 6 (28): 17-22.

Mota R D, Alonso S M, Ramírez N R y Cisneros P M A 2000 ¿Qué hay detrás de los parámetros productivos? Agronegocios en México, 37:26-27.

Mota RD, Martínez-Burnes J, Ramírez-Necoechea R, Trujillo OME, López-Mayagoitia A y Alonso-Spilsbury M 2001 Uso de oxitocina al parto y su efecto sobre el cordón umbilical y la muerte neonatal. Memorias del XXXVI Congreso Nacional AMVEC. Querétaro, Qro. p. 96.

Ramírez N R 1997 Perspectivas de la producción porcina en América Latina. Memorias del VII Congreso Latinoamericano de Veterinarios Especialistas en Cerdos. Río Cuarto, Córdoba, Argentina. pp. 113-121.

Ramírez N R, Mota R D, y Alonso-Spilsbury M 1998 Módulo básico de porcicultura artesanal. Memorias del XVI Congreso Panamericano de Ciencias Veterinarias. Santa Cruz de la Sierra, Bolivia. p. 322.

Ramírez N R, Mota R D, Alonso S M y Cisneros P M A 1999a Cálculo del flujo productivo. En: Numerología Porcina Ed. UAM-X. México D. F. pp. 24-32.

Ramírez N R, Mota R D, Alonso S M y Cisneros P M A 1999b Interpretación de parámetros. En: Numerología Porcina Ed. UAM-X. México D. F. pp. 1-23.

Ramírez N R, Mota R D, Alonso S M y Cisneros P M A 1999c Utilización del costo de alimentación para establecer la rentabilidad de una granja porcina. En: Numerología Porcina Ed. UAM-X. México D. F. pp. 58-62.

Ramírez N R y Mota R D 2000 La porcicultura en el D. F. y área conurbada. Foro de participación ciudadana: la ganadería y el desarrollo rural integral en México, opciones para el cambio en el subsector pecuario. UAM-X. Octubre. 8 pp. (sin publicar).

Suárez B 1995 La porcicultura de traspatio y su potencialidad. En: Kato L (ed.). La Producción Porcícola en México: Contribución al Desarrollo de una Visión Integral. Univ. Autónoma Metropolitana-Azcapotzalco. Univ. Michoacana San Nicolás Hidalgo. México. pp. 171-195.

Suárez B y Barkin D 1990 Porcicultura: Producción de Traspatio, Otra Alternativa. Centro de Ecodesarrollo, México D. F. 216 pp.

Trujillo O M E 1998 Control de la información. En: Sistema de Producción Animal. Cerdos. SUA. FMVZ, UNAM. pp. 3-4.

Vázquez P G, Robles C A y Berruecos J M 1972 Análisis de la relación entre número de parto de lechones nacidos y destetados en cuatro diferentes razas en clima tropical. Técnica Pecuaria México, 23:12-18.

 Received 14 September 2001

Go to top