Introduction Insulin-like growth factor-I (IGF-I) is a naturally occurring polypeptide produced in the liver, muscle and fat tissues. IGF-I is present in the circulating blood supply, usually bound with 1 of 6 binding proteins (IGFBPs). These binding proteins form stable complexes with IGF-I that act to maintain IGF-I in circulation, and additionally control the distribution, function and activities of IGF-I (Hossner et al. 1997). IGF-I influences growth and development during the postnatal growth period through endocrine, autocrine and paracrine mechanisms. IGF-I subsequently mediates the effects of growth hormone (GH), has localised effects on tissues at or close to sites of production, and also acts as an insulin-like metabolic regulator (Hossner et al. 1997). IGF-I and IGF-II, through their interactions with hormones and binding proteins, form part of a very complex system; many aspects of which are not completely understood. The concentration of circulating IGF-I can be measured relatively easily in blood plasma or serum with appropriate assay techniques (see Holly and Cwyfan Hughes 1994). As a result of its relative ‘ease’ of measurement compared with GH (i.e. serial measurements are not required), IGF-I was historically used as a substitute marker for GH in breeding experiments. Blair et al. (1988) demonstrated that divergent selection for IGF-I resulted in changes to both IGF-I concentrations and growth performance in rodents. Moderate estimates of heritabilities for IGF-I were later reported for cattle (Davis and Bishop 1991; Herd et al. 1995; Johnston et al. 2001, 2002) and pigs (Lamberson et al. 1995; Luxford et al. 1998a; Lahti et al. 2001; Cameron et al. 2003; Suzuki et al. 2004), implying utility for selection purposes in domestic livestock species. Single trait divergent selection lines for IGF-I are currently under development for cattle and sheep (Davis and Simmen 1997; Blair et al. 2002) but animal numbers are relatively low. While there are several studies that confirm circulating IGF-I concentrations are usually positively correlated with growth status in laboratory and livestock species, this is not always the case. It should also be noted that the majority of work reported on IGF-I has not been conducted on very juvenile animals, as is the case considered for trials here. In growing animals, Hossner et al. (1997) noted that increased growth in rats due to GH administration did not increase circulating levels of IGF-I, although IGF-I concentrations in hepatic and renal tissues were increased. Dunaiski et al. (1997) further showed that the infusion of Australian Journal of Experimental Agriculture, 2005, 45, 783–792 0816-1089/05/080783 10.1071/EA05048 © CSIRO 2005 K. L. Bunter A,C , S. Hermesch A , B. G. Luxford B , H-U. Graser A and R. E. Crump A A Animal Genetics and Breeding Unit, University of New England, Armidale, NSW 2351, Australia. B QAF Meat Industries, PO Box 78, Redlands Road, Corowa, NSW 2646, Australia. C Corresponding author. Email: kbunter2@une.edu.au Abstract. Insulin-like growth factor-I (IGF-I) is a naturally occurring polypeptide produced in the liver, muscle and fat tissues. It is known to be associated with growth and development during the postnatal growth period. Evidence for strong genetic correlations between juvenile IGF-I and performance traits would suggest this physiological measure would be useful as an early selection criterion. This paper reports estimates of genetic parameters from 9 trials where IGF-I was measured in juvenile pigs. All trials involved populations undergoing active selection for improved performance (e.g. efficient lean meat growth). Juvenile IGF-I was moderately heritable (average h 2 : 0.31) and influenced by common litter effects (average c 2 : 0.15). Genetic correlations (r g ) between juvenile IGF-I and backfat (BF), feed intake (FI) or feed conversion ratio (FCR) traits were generally large and positive: r g averaged 0.57, 0.41 and 0.65, respectively. Phenotypic correlations (r p ) between juvenile IGF-I and BF, FI or FCR were much lower (r p averaged 0.21, 0.09, and 0.15, respectively) as residual correlations between IGF-I and these performance traits were low, consistent with being measured at very different times. Correlations (genetic or phenotypic) between juvenile IGF-I and growth traits (e.g. lifetime daily gain or test daily gain) were relatively low, with average values within ± 0.09 of zero. Results from the trials reported here, and several physiological studies, indicate that information on juvenile IGF-I concentration can be used as an early physiological indicator of performance traits traditionally measured later in life. There is a clear role for juvenile IGF-I to facilitate pre-selection and more accurate selection of livestock for hard to measure traits, such as FCR, in pig breeding programs. Insulin-like growth factor-I measured in juvenile pigs is genetically correlated with economically important performance traits www.publish.csiro.au/journals/ajea CSIRO PUBLISHING