SciELO - Scientific Electronic Library Online

vol.49 issue3Phenotypic plasticity of Cuban Zebu sires evaluated in different Cuban regionsModeling of live weight per age in fattening bovines under a silvopastoral system with Leucaena leucocephala author indexsubject indexarticles search
Home Pagealphabetic serial listing  


Services on Demand




  • Have no cited articlesCited by SciELO

Related links

  • Have no similar articlesSimilars in SciELO


Cuban Journal of Agricultural Science

Print version ISSN 0864-0408On-line version ISSN 2079-3480

Cuban J. Agric. Sci. vol.49 no.3 Mayabeque July.-Sept. 2015




Short communication: U.S. dairy selection programs impact in Argentina


Comunicación corta: impacto de los programas de selección de ganado lechero estadounidense en Argentina



P. R. Marini,I R. G. López,II R. J. Di Masso,III

ICátedra de Producción de Bovinos Lecheros. Facultad de Ciencias Veterinarias - UNR, Ovidio Lagos y Ruta 33, 2170 Casilda, Argentina.

IIInstituto de Ciencia Animal – ICA. Cuba.

IIICátedra de Genética, Facultad de Ciencias Veterinarias – UNR, Argentina.




The aim of this study was to evaluate the milk index as an indicator of productive efficiency for dairy cows on pasture systems, compared with the usual indicator based solely on accumulated individual production. Age, total milk production and milk index from 127 dairy cows, classified by the Holstein Argentino Breeders Association (ACHA), were used, showing a cumulative production greater than 50,000 liters. Data were obtained from a commercial herd kept in the city of Totoras, province of Santa Fe, Argentina, during the period 1990-2009. The evidence confirm previous results showing that greater individual production at the end of the productive life of cows do not always guarantee a better result in terms of productive efficiency. The milk index, as an indicator of productive efficiency, allows a better assessment of the performance of dairy cows kept on pasture systems. It is concluded that this indicator has intrinsic advantages as it quantifies the level of adjustment between the genetic potential of the herd for milk production and the particular characteristics of the production system in which that potential is expressed.

Key words: milk index, genotype x environment interaction, artificial selection, imported semen.


El objetivo de este estudio fue evaluar el índice lechero como un indicador de eficiencia productiva para vacas lecheras en sistemas de pastos, comparado con el indicador usual que se basa solamente en la producción individual acumulada. Se utilizaron la edad, producción de leche total e indicador lechero de 127 vacas lecheras clasificadas por la Holstein Argentino Breeders Association (ACHA), que mostraron una producción acumulativa mayor a 50 000 litros. Los datos se obtuvieron de un rebaño comercial ubicado en la ciudad de Totoras, provincia de Santa Fe, Argentina, durante el período de 1990-2009. La evidencia confirma resultados anteriores que demuestran que una mayor producción individual al final de la vida productiva de las vacas no siempre garantiza un mejor resultado en la eficiencia productiva. El índice lechero, como indicador de la eficiencia productiva, permite una mejor evaluación del comportamiento de vacas lecheras en sistemas de pastos. Se concluye que este indicador tiene ventajas intrínsecas debido a que cuantifica el ajuste entre el potencial genético del rebaño para la producción de leche y las características específicas del sistema productivo en el que se expresa este potencial.

Palabras clave: índice lechero, interacción medio ambiente x genotipo, selección artificial, semen importado .




In a paper written as a contribution for the Symposium: Selection for milk yield, Hansen (1999) discusses the consequences of this procedure from the point of view of a geneticist, and argues that because of the global replacement of native populations of dairy cows all over the world, the consequences of selection programs in the United States would have evident international implications. North American Holsteins have been selected for increased body size for many years (Van Raden et al. 2010). Scores gathered by the Holstein Association USA (Brattleboro, VT) continue to place more favorable ratings on cows with larger body size through the use of body size composite, which is calculated from the 4 linear traits of stature, strength, body depth, and rump width (Holstein Association USA Inc. 2011). This statement seems to be applicable to dairy production in Argentina, taking into account that most of the semen used to inseminate Argentinean Holstein cows is from American origin. In 1997, Argentina was the second largest importer of bull semen in the world (Beltramino 1997). Argentine imported a total of 2.3 million semen doses in 2001 and 83% of them were from bulls of dairy breeds from the U.S. and Canada. As this situation has not change nowadays (CABIA 2008), it allows speculating that the genetic progress observed in the countries of origin of these bulls moved to a large extent to the herd of Holstein cows in Argentina, a concept that is generally shared by importers, sellers and advisers. The consequences of such assertion clash with the purposes stated in a totalizing vision of national milk production (Molinuevo, 2005) that emphasizes the adjustment that must exist between the genetic potential of a productive species and characteristics of the environment in which that species should express this potential. This mismatch is reflected on a deterioration of the reproductive performance of cows of high productive performance, when reared in an environment not able to fulfill their nutritional requirements. To characterize more accurately the performance of dairy cows in these grazing systems, a new indicator called milk index (MI) was defined (Marini and Oyarzabal 2002ab). Milk index includes the total milk production accumulated by one cow over all the lactations by the time required for the same cow to achieve this production (MI: total milk production in liters/number of days required to produce them, i.e.: age in days at the end of the last lactation minus age in days at the first calving). Its use allows the comparison among cows with different number of lactations, including the impact of eventual reproductive failures not considered by total milk production, and the traditional value is used as productive indicator.

The main aim of this study was to empirically test the hypothesis that the use of semen of Holstein bulls selected in the United States resulted in a mismatch of local populations commonly used for milk production in pasture systems, and to discuss the implications of the practice of inseminating cows with semen from bulls selected in a different productive



Data from 127 dairy cows classified by ACHA (Argentine Holstein Breeders Association) were used, as belonging to a particular category because of their cumulative production of more than 50,000 liters. They were obtained from a commercial herd establishment located in the city of Totoras, Santa Fe, Argentina, during the period 1990-2009. Milk index values for each cow were plotted against average milk production per lactation of the same cow. The property comprises official milk control Totoras Rural Society, official agency No. 13 official entity and presents management guidelines (health, food and technical assistance) which places in these areas above the overall average of that basin. Food is basically pasture (alfalfa) with supplementation (corn grain, corn silage and rolls) provided in different proportions according to the seasonal availability of alfalfa pastures. Gynecological examination was periodically performed and the cows were inseminated with semen from American and Canadian origin. As the relationship between milk index and average milk production per lactation, within each group, was consistent with a linear model, characterized by the value of the two parameters (slope and intercept) of the linear function, they were estimated by linear regression. The group effect on the values of both estimates was assessed by mean of an analysis of covariance. Cows belonging to each of the mentioned groups were also characterized in terms of the values of the following traits: AGE (cow age, in days), PROD1 (total milk production, in liters), PROD2 (daily milk production, in liters per day), MI (milk index, in liters per day), CALV (number of calving), SERV (total number of services per pregnancy), INT (calving interval, in months), and FAT (milk fat content, in %). The group effect on each response variable was assessed by means of a one way analysis of variance, followed by Bonferroni multiple comparisons test as in the case of using the parametric alternative and the nonparametric variable number de calving (Kruskal-Wallis rank test) followed by Dunn multiple comparisons tests as in the case of using the nonparametric alternatives.



Visual inspection of the resulting scatter diagram (figure 1) allowed to identify four groups of animals which were characterized according to the following description:

Group 1: Cows similar to those belonging to a commercial dairy farm in a pasture system.

Group 2: Cows with high milk production within the trend showed by Group 1

Group 3: Cows with similar variation in milk index values but displaced to the right, i.e. with higher milk production per lactation

Group 4: Cows with high productivity.

A statistically significant difference between the slopes of the regression lines (F=2.87, P=0.0395), prevented to compare their intercept values (table 1). This result revealed differences in the pattern of change of milk index values in terms of increased average daily production of milk in each of the four groups of cows.  As table 2 shows, cows of Group 1, with the lowest milk index values, can be characterized from a productive point of view as those with the lowest average milk production per lactation, the lowest milk production per day and the highest milk fat content. From a reproductive point of view, they have less number of services per pregnancy and lower calving interval. Cows in Group 4 showed the highest average milk production per lactation, the highest milk production per day and the lowest milk fat content. They were also those that required more services per pregnancy and showed the highest calving interval. This group includes the youngest cows, and although it presents, in average, milk index values significantly higher than Group 1, it does not differ from cows located in Group 2 and Group 3 in terms of MI values. Cows in Group 2 did not differ from those located in Group 1 either in age and number of calving or in the mean values of the other reproductive traits. So, their high milk index values did not demonstrate differences in reproductive traits, but in their best productive performance due to a higher average milk production per lactation and per day, without significant differences in milk fat content. Cows in Group 1 represent those with normal production levels in dairy farms of the region, which do not affect their reproductive performance when reared on pasture. Cows in Group 2 would represent the type of cow to be expected if the productive level of those in Group 1 is increased by improving basically the nutritional environment. In contrast, cows in Group 3 present the same average milk index values than those in Group 2, but their increased milk production was achieved at the expense of deteriorating their reproductive performance, in correspondence with the results of Henández et al. (2014). Cows in Group 4 would be the most extreme case of this trend observed in the information collected, with higher production, greater reproductive impairment and equal rate of milk. With the current evaluation criteria that prioritize the productive aspects, cows belonging to Group 4 first, and then Group 3 would be best to present both the best productions of average daily milk per lactation. But when they are evaluated based on their rates of milk, and they are no longer better than those in Group 2, with lower production values but better reproductive rates exhibited the same rate of milk. An index of 13 liters of milk/day can be obtained with 6,800 (Group 2), 7,800 (Group 3) and 9,400 (Group 4) liters and its value reveals reproductive impairment concomitant with better growth performance. On the other hand, the same reproductive performance may be accompanied by productions of 5,200 (Group 1) or 6,800 (Group 2) liters per lactation, and, in this case, the difference in terms of production shows a better milk index value (10 vs. 13 liters / day). These data confirm previous results showing that increased individual production at the end of the productive life of the cow does not always guarantee better production efficiency (Marini and Oyarzabal, 2002 a b). In these production systems based on direct grazing, there would be a limit on the expected production of a cow above which reproduction suffers. Since the cow to produce milk must be reproduced, there is a clear antagonism to be taken into account when deciding the criteria to be applied in improving such systems. Therefore, it is important to have information on the implications of directional selection for production, first, for using the selection criteria that do not depress other characters such as reproductive, and, second, for contributing to the discussion on how to determine which is the most suitable and profitable grazing system leading to greater sustainability of such systems. Figure 1 shows  a clear evolution of the animal that was included to these systems in Argentina since the beginnings of the biotypes in 1992, with the massive importation of semen of American and Canadian origin. The quest for greater individual production massively imposed as a general criterion for all drums. The use of milk yield as a selection criterion allowed a significant genetic progress in the countries that lie at the forefront of dairy breeding. In the U.S., for example, the accumulated genetic progress of the bulls used in artificial insemination increased in 37 kg per year in the 60´s, 79 kg per year in the 70´s, 102 kg per year in the 80´s and 116 kg per year in the early 90´s (Hansen 1999). Genetic improvement is one, but not the unique, tool for achieving an efficient and competitive dairy production. In developed countries, the association of selection objectives, adequate to their production conditions with the successful use of genetic resources, has allowed to accumulate a genetic progress for milk production from 0.8 to 1.5% per year. In Argentina, genetic improvement of dairy cattle has been interpreted as synonymous of the use of imported semen. This approach ignores that the biological and economical efficiency of dairy farms depend not only on the higher production per lactation, but also on a set of components like cow longevity, its reproductive efficiency  and the efficiency of the rearing process of heifers for replacement. The existence of genotype-environment interactions made evident the necessity of restricting the use of imported semen for the insemination of cows kept under intensive confined systems with a high level of supplementation that ensure the fulfill of their nutritional requirements. For the maintenance of traditional outdoor production systems on pasture it is necessary to use semen from bulls which had been selected under similar local environmental conditions, it is usually done in beef production systems.

In this context, the milk index is presented as an indicator of productive efficiency that enables a comprehensive assessment of the performance of dairy cows kept in grazing systems because it comprises the degree of fit / mismatch between the genetic potential of producing each individual and the characteristics of the production system in which this potential must be expressed.



The authors would like to acknowledge MV Gustavo Magnano for contributing data to this study.



Beltramino, F. E. 1997. “Eficiencia de razas lecheras”. In: Temas de Producción Lechera, (ser. Publ. Misc., no. ser. 84), INTA, pp. 59–65.

CABIA. 2008. Cámara Argentina de Biotecnología de la Reproducción e Inseminación Artificial. , Available: < estadistica.aps> .

Hansen, L. B. 1999. “Consequences of selection for milk from a geneticist`s view point”. J. Dairy Sci., 83: 1145–1150.

Hansen, L. B. 2005. Vacas lecheras funcionales. EE.UU: Universidad de Minnesota, 3 p.

Hernández, A., Ponce de León, R., García, L. R., Marini, P. R., Guerra, D., Padrón, Y., García, S. M., López, M., González, S., Quiñones, D., López, O. & Pacheco, R. A. 2014. “Relación genética entre la producción de leche, la reproducción y la longevidad en vacas Mambí de Cuba”. In: XV Jornadas de Divulgación Técnico Científicas en Ciencias Veterinarias y II Jornada Latinoamericana, pp. 191–192.

Holstein Association USA Inc. 2011. Holstein type-production sire summaries. Brattleboro, VT: Holstein Association USA Inc.

Marini, P. R. & Oyarzabal, M. I. 2002a. “Patrones de producción en vacas lecheras. 1 Componentes de la producción y sus características según nivel de producción”. Rev. Arg. Prod. Anim., 22 (1): 29–46.

Marini, P. R. & Oyarzabal, M. I. 2002b. “Patrones de producción en vacas lecheras. 2 Descripción de la vaca promedio y estimación de los ingresos según categorías de producción”. Rev. Arg. Prod. Anim., 22 (1): 47–60.

Molinuevo, H. A. 2005. Genética bovina y producción en pastoreo. INTA, 348 p.

Vanraden, P. M., Cole, J. B., Tooker, M. E. & Cooper, T. A. 2010. Genetic base changes for January 2010. , Available: <> .



Received: January 23, 2014
Accepted: March 16, 2015



P. R. Marini, Cátedra de Producción de Bovinos Lecheros. Facultad de Ciencias Veterinarias - UNR, Ovidio Lagos y Ruta 33, 2170 Casilda, Argentina. Email:

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License