Dynamic analysis of the body weight of five maternal strains of breeding hens for free-range chicken production

Growth is a complex biological process which is affected by both genetic and environmental factors, as well as the respective interactions between these two. Body weight has traditionally been the variable of choice for its quantification. The significance of this process in domestic animals is of economic interest and is not restricted to those species destined for the production of meat in which the marketable product - the carcass - is a direct result of the same, but also to those used in other types of production, such as milk, wool, or eggs, since, in all these cases, the animal must necessarily pass through a period of growth before reaching productive stage. A second aspect related to growth is that the overall efficiency of any of these processes is affected by the mature size of the breeding populations.

These are some of the reasons why the growth has been one of the primary objectives of selection in animal production and one of the most widely used. The adjustment of the pattern of body weight variation based on time makes it possible to summarize the information contained in multiple weight-age points in only two parameters of biological interest: The asymptotic size and the maturation rate, which describe the dynamics of the process. The potential of dynamic growth studies based on the adjustment of mathematical non-linear models to longitudinal size-age data, in relation to the possibilities offered by the estimate values of these parameters with meaning biological to enable more detailed quantitative comparisons of growth patterns has led to view this particular type of analysis as a new paradigm (Brisbin et al. 1987) to the extent that the paradigms provide referential mental patterns that allow to redefine the modalities for approaching a particular problem (Gochfeld, 1987). In some productive situations, however, certain growth patterns may be more desirable than others (Tallis, 1968), and this depends on both of the species under study and the system of production in which the species is reared. Any advantageous pattern can, in principle, be described in terms of the above two parameters:  Asymptotic size and maturation rate. In meat poultry operations, for example, a desirable pattern should aim to achieve early optimal growth in birds with a high rate of protein deposition, accompanied by a later decline in growth rate with low fat deposition (Barbato, 1991). This work aims to characterize in a dynamic way the growth of five maternal strains of breeding hens used by INTA (Bonino, 1997; Bonino & Canet, 1999) for the production of free-ranger chickens on the basis of the value of the two biological parameters that define the shape of the growth curve: Asymptotic size (A) and maturation rate (k), estimated by non-linear adjustment of the body weight - chronological age longitudinal data.

Birds of five maternal strains (A, E, CE, DE and ES) used in the production of free-range chickens were evaluated. The birds were raised in a shed, together, subject to common management up to 12 weeks of age, when they were separated by breed and moved to the final pens (n = 100 birds by breed), where they remained until 50 weeks of age. Once a week, between birth and 50 weeks of age, the body weight of a random sample of 50 birds of each breed was recorded with 1 gram approximation. The longitudinal body weight - chronological age data were adjusted using the sigmoid model of Gompertz (Fitzhugh, 1976):

                                                           W (t) = A exp (-b exp (- k t))

Where:

W (t) = body weight (g) in time t
A = asymptotic body weight (average weight as t tends to infinity)
b = position parameter position, without any biological significance that adjusts for those cases that t is different from zero
k = maturation rate (speedy of approach to value A)
t = time in weeks

The growth curves adjustment was carried out by non-linear regression using an iterative method based on the Marquardt algorithm and comparison thereof was performed using the statistical program Graph Pad Prism, version 2.0. The advantage of this adjustment was evaluated from the convergence in a solution, the value of the non-linear determination coefficient (R²), and the randomness of the residual (runs test- Sheskin, 2000).

The following table summarizes the values (average ± standard error) of estimators of the parameters of asymptotic size (A) and (k) maturation rate corresponding to five breeds.

Significant differences among breeds were observed in the joint analysis of the trajectories of growth curves (F = 31,36, P < 0,0001 - A > DE = CE = ES > E). Although, according to the known negative association between mature size and speed of approach to this size, the breed with smaller asymptotic size (Breed E) presented the highest maturation rate; the remaining did not show the expected behavior. Breed A, being of larger mature size, presented a smaller average k value than breed ES, of less adult body weight. Breeds DE and CE showed a lower maturation rate for body weight that Breed ES, although they do not differ significantly in body weight in the growth asymptote.

The results show that, in terms of body weight, the evaluated breeds show differences in their growth patterns. These differences manifested themselves not only in the asymptotic size values, but also in the speed with which it is reached this size, thus generating different trajectories on their way to mature body weight and highlights the existence of an at least partially independent genetic basis for asymptotic size and maturation rate of maturation.

Thanks to Jose Acevedo and Emanuel Acosta for their cooperation in the gathering of data for this work.

Barbato G. 1991. Genetic architecture of growth curve parameters in chickens. Theoretical and Applied Genetics 83:24-32.

Bonino MF. 1997. Pollo Campero. Protocolo para la certificación. INTA. EEA Pergamino.

Bonino M & Canet ZE. 1999. El pollo y el huevo campero. INTA.

Brisbin IL, Collins CT, White GC, McCallum DA. 1987. A new paradigm for the analysis and interpretation of growth data: the shape of things to come. Auk 104:552-554.

Fitzhugh HA. 1976. Analysis of growth curves and strategies for altering their shape. Journal of Animal Science 42:1036-1051.

GraphPad Software, San Diego, California, USA, www.graphpad.com.

Gochfeld M. 1987. On paradigms vs. methods in the study of growth. Auk 104:554-555.

Sheskin DJ. 2000. Handbook of parametric and nonparametric statistical procedures. Chapman & Hall. USA.

Tallis GM. 1968. Selection for an optimum growth curve. Biometrics 24:169-177.