CROSSBREEDING CATTLE IN LATIN AMERICA

F.E. Madalena 1/

The economic efficiency of cattle production may be improved by the appropriate choice of breeds, mating system and selection of individuals within breeds, to obtain increased product yield or quality per unit of input resources (land, labour, capital).

Experimental results on the comparative performance of cattle breeds and crosses in tropical Latin America indicate that, in general, crossbreds outperform purebreds both for beef (Madalena, 1977; Hernandez, 1981; Plasse, 1981, 1983) and for dairy production (Vaccaro, 1979; Wilkins et al. , 1979; Muñoz and Deaton, 1981; Madalena, 1981; Madalena et al. , T983b; De Alba and Kennedy, 1985). Crossbreeding allows exploitation of heterosis, maternal effects and complementarity between breeds but the appropriate strategy depends on the characteristics of the production system considered.

1. PRODUCTION SYSTEMS

Tropical cattle production systems in Latin America were described by several authors (Plasse, 1976; Wilkins et al., 1979; Paladines, 1980; Madalena, 1981; Ruiz, 1982; Cubillos, T9"82; Sere and Vaccaro, 1984; CIAT, 1985). In general, beef cattle herds are kept in ranches and dairy cattle in smaller properties, but there is also a wide range of dual purpose systems. Coarse pastures/roughages limit nutrient intake, aggravated by periods of drought or flooding. Other constraints to cattle performance are: mineral deficiencies, incidence of diseases, ticks, torsalo grubs and gastrointestinal parasites, heat, humidity and solar radiation. Natural service is common, controlled matings and artificial insemination being practised at a minority of farms. Socio-economic constraints such as absentee ownership and low education level of rural populations are important background factors causing poor farm and herd management. It should be recognized that wide variations exist between and within regions, with many examples of good farming based on modern husbandry techniques.

2. MATCHING CATTLE GERMPLASM TO PRODUCTION SYSTEM

Latin American cattle breed resources may be conveniently grouped into four classes:

1. the founder Criollo naturalized initial populations, now mostly graded up;

2. zebu breeds;

3. the modern European breeds selected for high performance in temperate regions; and

4. new breeds derived from crosses between European and adapted breeds, like the Santa Gertrudis, Canchim and Ibage for beef production and the Jamaica Hope, Pitangueiras and Siboney for milk production, to mention some examples.

Modern European breeds may be utilized only in the more intensive production systems with no important climatic constraints, but are totally unfit for the harsher environments. Criollo and zebu breeds are adapted to harsh environments because of their heat tolerance, low metabolic rate and disease and parasite resistance but have comparative low performance in improved environments. For a wide range of intermediate environments, complementarity between higly productive and adapted breeds results in superior overall performance of crossbreds.

Our own results on crossing red and white Holstein-Friesian (HF) x Guzera (G) in Brazil may be used as an example of this situation (Madalena, Lemos, Teodoro and Monteiro, in preparation). Milk yields, calving intervals and milk yields per day of calving interval of six crossbred groups (grades) are shown in Figures 1, 2, 3 The six HF grades were: 1/4, 1/2, 5/8, 3/4, 7/8 and > 31/32 or HF. The halfbreds were f₁ out of G dams by HF sires. The 1/4 and 3/4 were first backcrosses of F₁ dams to, respectively, G and HF sires. The 7/8 were second backcrosses to HF sires, and the 5/8 were obtained by inter se matings of 5/8 sires and dams.

Animals were produced at an experimental farm and distributed to 66 cooperator farmers at mean age 22 months, to measure dairy performance under a wide range of commercial practices. Cows were milked in the presence of the calf, which suckled after milking, according to the generalized practice in the region. However, no milk was intentionally left for the calves on recording days. Farms were grouped into high and low management level classes for purpose of analysis. Figure 1 is based on 921 observations and Figures 2 and 3 on 699.

Figure 1 Lactation milk yields for six Holstein-Friesian x Guzera grades at high (*) and low (°) management levels.

Cunningham (1981) described the situation depicted in Figures 1 and 3 as a double interaction of additive breed difference and heterosis by environment, the additive effect becoming more important and heterosis less important as the level of environmental stress is reduced. Results from another farm with mean milk yield of 9.8 kg per day of calving interval indicated no differences in this trait for grades 3/4, 7/8 and HF in HF x Gir crosses (Madalena et al., 1983a). Cuban results are in agreement with this conclusion (Ponce de Leon et al. , 1982).

Figure 2 Calving intervals for six Holstein-Friesian x Guzera grades at high (*) and low (°) management levels.

Figure 3 Milk yield per day calving interval for six Holstein-Friesian x Guzera grades at high (*) and low (°) management levels.

Another example of genotype x environment interaction - perhaps an obvious one - is shown in Figure 4, where tick (Boophilus microplus)field burdens are shown for heifers of the six HF x G grades at 12 different occasions (Lemos et al ., 1985). At the low levels of infestation, differences between gradles in tick resistance were small, but they amplified at higher infestation levels. The economic values of the genetic tick resistance of zebus and Criollos (Ulloa and De Alba, 1957) depends on level of infestation, and would become irrelevant should ticks be controlled by pasture spelling (Sutherst et al., 1979) or, in the future, by vaccination (R.W. Sutherst, personal communication).

Figure 4 Average tick count per heifer for six Holstein - Friesian: Guzera grades at twelve assessment occasions.

3. CROSSBREEDING STRATEGIES

The HF x G results presented above belong to an experiment designed to compare the following crossbreeding strategies (Madalena, 1981):

1. Grading up to HF (represented by the HF group).

2. Creating a new breed (represented by the 5/8 group).

3. Crisscrossing, or rotation in each generation of HF and G sires.

4. Modified crisscrossing, repeating the HF sire breed for two generations, followed by G sires.

The latter procedure was suggested by Madalena (1981) to maintain the crossbred population at higher European grades (3/7, 5/7 and 6/7) than would be possible by crisscrossing (1/3 and 2/3). Mean performances for these two strategies were estimated from predicted means of the resulting grades under a breed additive difference and heterosis model (Dickerson, 1973). The results for milk yield per day of calving interval are shown in Table 1. F₁ were taken as a reference because they had the highest performance of all six groups (Figure 3).

Table 1 MEAN MILK YIELD PER DAY OF CALVING INTERVAL FOR SEVERAL CROSSBREEDING STRATEGIES OF HOLSTEIN-FRIESIAN (HF) X GUZERA (G), RELATIVE TO F₁- PERFORMANCE

Does not consider the purebred herd necessary to produce the F₁.Continuous production of F₁ heifers would not be practical on a regional scale, but it might be a good commercial proposition for individual farms in specific cases (Madalena, 1981).

Crisscrossing repeating the HF sire showed the second highest performance at both management levels (Table 1). Although the practical rule to apply this scheme is very simple, it requires the mating of females with a grade higher than 3/4 HF to the G bull and the other females to the HF bull, most farms would lack the organization required to keep track of which females should be mated to each sire breed. The scheme requires keeping at least one bull of each breed, which would not be economical for small herds, so it is better suited for artificial insemination in this case.

Dairy farmers unable to organize rotational crossing would still have two ways of maintaining the herd at intermediate grades:

1. Periodic switching of the sire breed (Roger, 1973). This is the prevailing practice at dairy farms in southeast Brazil (Madalena, 1981). it has the disadvantage of producing a high proportion of extreme genotypes, with too much or too little zebu breeding.

2. Use of crossbred bulls. Poor performance should be expected from inter se matings of unselected crossbred bulls and cows, as those used in our experiment. Negative effects of heterosis breakdown should be counteracted by selection for milk yield. Conventional progeny testing (using elite herds) would be quite possible in many Latin American countries, so that bulls for natural service could be produced by artificial insemination of the better cows with semen of proven bulls. In my opinion, difficulties for the implementation of a scheme of this sort lie more in poor organization of public institutions than on other factors.

Purebred European cattle may be the preferred option for the more intensive systems, particularly when climatic stress is attenuated by high altitude. Based on the results mentioned above, it would appear that purebred HF may be recommended for systems capable of sustaining lactation milk yields of at least 4000 kg, 10 kg per day of calving interval and calf mortality of 15 percent or less up to one year of age.

The elements for deciding on breeding strategy for beef cattle are quite different from those considered for dairy cattle. Practical problems to implement rotational crossing should not be very important as separate breeding herds may be kept for each sire breed (Madalena, 1977). On the other hand, reproductive efficiency of non-adapted European bulls may be seriously impaired in natural service under extensive conditions (Table 2), so the introduction of genes from these breeds would require the use of crossbred bulls (or artificial insemination when possible). However, heterosis breakdown seems to be less important for reproductive and growth traits than for milk yield and lactation length. Plasse (1983) indicated that zebu-Criollo rotational crossing or composite populations including also other European breeds are promising alternatives, although not enough information is available yet for a final comparison of breeding strategies.

Table 2 REPRODUCTIVE EFFICIENCY OF BULLS OF SEVERAL BREEDS IN NATURAL SERVICE TO NELORE FEMALES IN BRAZIL (from Razook et al., 1985)

^1/ Includes some artificial insemination.

REFERENCES

^1/ Centro Nacional de Pesquisa - Gado de Leite, Caixo Postal 151, Coronel Pacheco 10 MG, Brazil.