Re-evaluating nutrition models to predict calf growth M. I. Marcondes1, T. E. Da Silva2, J. H. C. Costa3 1William H Miner Institute, Chazy, NY 2Department of Animal and Veterinary Sciences, University of Vermont Introduction: Why nutrition models for calf growth matter The rearing of calves and heifers represents one of the largest investments in dairy operations, accounting for approximately 15–20% of the total cost of producing milk (Heinrichs, 1993; Van Amburgh et al., 1998). These animals do not generate revenue until first calving, which makes the efficiency of their growth programs essential for farm profitability. Inaccuracies in predicting growth may translate into not optimized feeding programs that either oversupply or undersupply nutrients, both of which can be costly. Overfeeding leads to excess rearing expenses and metabolic problems, while underfeeding delays age at first calving (AFC), prolonging the non-productive period. Nutrition models have been developed aiming to match supply with requirements by predicting energy and protein needs for target growth rates. However, these predictions rely heavily on the accuracy of the underlying growth equations. If models underestimate or overestimate growth potential, heifers may reach breeding size either too late or in suboptimal condition, which affects reproduction, longevity, and lifetime productivity (Kusaka et al., 2023). Importantly, growth in early life has programming effects that extend into lactation. For example, calves that achieve higher preweaning average daily gain (ADG) consistently produce more milk in their first lactation (Soberon et al., 2012; Van De Stroet et al., 2016). This demonstrates that growth is not simply a cost to be minimized, but a process with long-term biological and economic implications, and highlights the need for accurate and adaptable prediction models. Biological basis of calf and heifer growth Growth in dairy calves and heifers occurs in distinct developmental phases, each with unique nutritional priorities. During the preweaning phase, rapid skeletal and organ development predominates. Increased nutrient supply at this stage, particularly through greater liquid feed allowances, has been linked to earlier puberty, increased milk yield, and higher economic returns (Davis Rincker et al., 2011; Soberon et al., 2012). In the post-weaning to puberty phase, skeletal growth accelerates, but the mammary gland is particularly sensitive to nutritional balance. Studies have shown that excessive energy intake during this window can increase fat deposition within the gland and impair parenchymal tissue development, ultimately reducing future milk yield (Sejrsen et al., 1982; Sejrsen & Purup, 1997). After puberty, growth increasingly reflects fat deposition relative to frame size. Proper ration formulation is required to support breeding readiness—commonly targeted at 55% of mature BW—without creating over-conditioned heifers that are prone to dystocia or metabolic issues (Heinrichs, 1993; Hoffman, 1997). Finally, during gestation through calving, nutrients are partitioned to support the fetus growth. Inadequate supply compromises calf development, while excessive energy predisposes heifers to calving difficulties and metabolic diseases. These biological realities are captured imperfectly by models such as NASEM (2021), which predict nutrient requirements based on BW and target ADG. While such models have provided a strong foundation, they often fail to account for structural growth (e.g., height, body proportions) or the long-term programming effects of early nutrition, which limits their ability to guide precision feeding strategies across diverse environments. Protein requirements considerations Microbial protein for calves For decades, nutrition models have assumed that preweaned calves derive virtually all of their metabolizable protein from dietary and endogenous sources, largely ignoring microbial crude protein (MCP) as a contributor. This assumption stems from the fact that the esophageal groove shunts most of the liquid diet directly to the abomasum, bypassing the rumen. However, recent research challenges this view, demonstrating that microbial protein production begins much earlier in life than previously recognized. Pinheiro et al. (2025) quantified the extent of milk replacer leakage into the rumen and measured MCP synthesis in preweaned dairy kids. On average, 56% of the milk replacer was absorbed into the rumen, providing a substrate for microbial fermentation. This process generated volatile fatty acids, ammonia, and microbial biomass, with MCP synthesis averaging 1.95 g/d. Although the average MCP synthesis was modest (~1.95 g/d), its nutritional significance becomes clearer when expressed relative to crude protein (CP) intake. The kids in the study consumed between 28.3 and 56.7 g CP/d, meaning that MCP represented on average a small fraction of their protein supply (~4.5%). However, the variability was substantial: MCP ranged from as low as 0.625 g/d to as high as 4.0 g/d. Consequently, the proportion of dietary CP (from milk replacer) converted into MCP spanned from 1.1% up to 14%. At the lower end, 1.1% supports the traditional assumption that MCP contribution in preweaning ruminants is negligible. Yet at the upper end, 14% challenges this notion and suggests that microbial activity may represent a meaningful and previously underestimated source of protein during early life. This degree of variation highlights both the potential importance of MCP and the limitations of current requirement models that ignore it. Unfortunately, the factors driving such variability in MCP production remain unclear, and future research is warranted to identify dietary, physiological, or microbial drivers that could allow nutritionists to harness this early microbial contribution to enhance preweaning nutrition. Complementary evidence from Pinheiro et al. (2025) shows that calves fed higher milk replacer allowances exhibited improved nitrogen retention, increased nutrient flow to the gastrointestinal tract, and enhanced tissue hypertrophy, including skeletal muscle development. This suggests that although MCP yield may be modest in absolute terms, early microbial activity interacts with systemic metabolism, influencing nitrogen utilization efficiency and muscle protein synthesis. These insights have important implications for re-evaluating protein requirements in calves. Current requirement models (NRC, 2001; NASEM, 2021) do not account for MCP contribution in preweaning diets, essentially treating the calf as a monogastric animal. Yet the demonstrated leakage of milk replacer into the rumen and measurable MCP synthesis call for adjustments in how we model metabolizable protein supply during early life. Even small contributions of MCP may reduce the reliance on dietary protein or interact with growth signals such as insulin-like growth factor-1 (IGF-1), which has been linked to muscle development in calves and kids fed higher liquid allowances (Pinheiro et al., 2025). Overall, the emerging picture is that microbial protein synthesis starts earlier than traditionally recognized, albeit at a lower magnitude than in post-weaned animals. Integrating this contribution into calf protein requirement models could improve the accuracy of predictions and better align nutritional strategies with the biological reality of early rumen development. Starter crude protein for calves As the milk allowance is reduced during the preweaning period, calves become increasingly dependent on starter to supply the necessary nutrients for growth. Importantly, the rise in nutrient demand is not uniform: requirements for energy increase in a greater proportion than requirements for protein. This has led to the hypothesis that starter crude protein (CP) concentration could be gradually reduced over time without compromising performance, provided that intake is sufficient. Results from requirement models support this idea, showing that the efficiency of utilizing metabolizable protein (MP) and metabolizable energy (ME) from starter is markedly lower than from milk, with estimates of 44.4% for MP and 41.2% for ME from starter compared with 71.9% and 57.6%, respectively, from milk (Marcondes and Silva, 2021). Thus, energy capture tends to become the primary limiting factor once milk is stepped down, and excess protein supplied in this context may be used inefficiently. In an attempt to align dietary supply with these changing requirements, we recently evaluated a decreasing-CP starter strategy, which offers calves a higher CP concentration early in life (when starter intake is minimal) and progressively reduces CP content as milk is withdrawn and starter intake increases (Silva et al., 2025). Contrary to our expectations, the decreasing CP approach did not enhance preweaning growth. Instead, calves on the fixed 18% CP starter exhibited greater weight gains and feed intake, while those on the decreasing CP program had lower starter intake and reduced fecal output near weaning. Nitrogen-use efficiency improved with the decreasing CP program, and fecal N excretion was reduced, suggesting potential environmental benefits; however, these came at the expense of calf growth and robustness of performance. Given these findings, why have some nutritionists moved in the opposite direction, formulating calf starters with 24–25% CP? The rationale stems from two factors. First, the 2021 NASEM model shifted calf protein requirements from a crude protein to a metabolizable protein framework. In this system, MP needs are driven by target daily gain, and the efficiency of using MP varies by source: ~0.95 for milk proteins, ~0.75 when liquid and starter are fed together, and ~0.70 once the rumen is more developed (NASEM, 2021). Second, microbial protein supply only becomes a major contributor once starter intake exceeds ~1.3 kg/d, which often occurs after weaning. At typical preweaning starter intakes, microbial N is still limited, so nutritionists usually hedge by increasing CP density to ensure that essential amino acid requirements (particularly lysine and methionine) are met from plant proteins. Several trials illustrate the outcomes of this high-CP approach. Stamey et al. (2012) compared a 19.6% CP starter with a 25.5% CP starter in calves receiving either conventional or enhanced milk replacer. Within the enhanced program, calves fed the higher-CP starter showed greater starter intake around weaning, a tendency for higher average daily gain, and heavier body weights at 8–10 wk. In a follow-up study, Lanier et al. (2021) found that increasing starter CP from 21.5 to 26% altered tissue composition (more lean, larger visceral organs) under a high milk replacer program, but did not increase body weight gain per se. More recently, Yousefinejad et al. (2021) reported that increasing starter CP from 18 to 22% improved average daily gain, weaning weight, and feed efficiency, regardless of the proportion of rumen-undegradable protein. Collectively, these studies explain why 24–25% CP starters are attractive in accelerated growth programs: they can support lean gain and smooth the transition at weaning, especially when milk supply is reduced early and starter intakes remain modest. It is important to emphasize that energy is usually the most limiting nutrient for calves. Therefore, the optimal crude protein concentration in the starter should be considered in relation to both milk allowance and starter intake. In practice, higher CP levels tend to elicit stronger responses primarily when calves are receiving greater amounts of milk. Nevertheless, evidence is inconsistent, and our own work suggests caution. Our trial demonstrated that although decreasing CP increased nitrogen efficiency, reducing CP during the step-down phase compromised growth—likely due to amino acid imbalances, lower efficiency of using starter-derived MP, and intake dynamics near weaning. NASEM (2021) does not prescribe a universal CP level; instead, it emphasizes formulating to meet MP and essential amino acid requirements according to growth targets and intake predictions. From a practical standpoint, a CP level of 20–22% is sufficient in most conventional milk programs. Higher CP (22–24%, occasionally up to 25%) may be justified in aggressive weaning systems with high growth targets and limited starter intake, but only if amino acid balance is ensured. Importantly, as starter intake rises and microbial protein supply stabilizes, phasing CP downward post-weaning is biologically sound and environmentally responsible. In summary, the rationale for 24–25% CP starters is strongest in aggressive programs where calves consume little starter preweaning and amino acid supply from solid feed is at risk of being limiting. However, responses are variable, and excessive reliance on CP percentage rather than amino acid balance may lead to inefficiencies. Current evidence does not support blanket recommendations for very high CP starters. Instead, formulation should focus on supplying metabolizable protein and essential amino acids in synchrony with energy, aligning with NASEM (2021) principles. Further research is needed to optimize amino acid-balanced strategies that separate “more protein” from “the right protein at the right time.” Metabolizable protein and energy for postweaned heifers In postweaned dairy calves, the balance between metabolizable protein (MP) and metabolizable energy (ME) intake is critical for guiding tissue partitioning and mammary development. A higher MP:ME ratio has been associated with improved deposition of mammary parenchymal tissue while limiting undesirable fat accumulation in the mammary gland, particularly when heifers are allowed higher rates of gain. Albino et al. (2015) demonstrated that Holstein heifers gaining 1.0 kg/d maintained adequate mammary development only when diets supplied more than 43 g of MP/Mcal of ME, whereas lower MP:ME ratios led to greater fat deposition in the gland. This aligns with recent NASEM (2021) recommendations for heifers with BW > 125 kg, which propose estimating the minimum MP requirement relative to energy as: MP (g/Mcal ME) = 53 − 25 × (BW/Mature BW). For a 200-kg heifer with a mature BW of 700 kg, the equation predicts ~46 g MP/Mcal, which is consistent with Albino’s threshold. However, by 400 kg BW the equation allows the ratio to decline to ~39 g/Mcal. Our results and those of Albino et al. (2015, 2017) suggest that such low values are not adequate when heifers are gaining around 1 kg/d, and a practical minimum of ≥43 g/Mcal ME should be maintained to avoid excessive fat deposition in the mammary gland. Similarly, Weller et al. (2016) and Albino et al. (2017) confirmed that high nutrient intake levels stimulated systemic IGF-1 but also promoted lipogenesis in the mammary gland, indicating that energy supply in excess of available MP can shift nutrient partitioning toward adipose tissue rather than functional parenchyma. These findings support the idea that mammary gland development is more sensitive to the protein-to- energy balance than to overall growth rate alone (Silva et al., 2002; Albino et al., 2015). In fact, Silva et al. (2002) showed that body fatness, rather than body weight gain per se, was the stronger predictor of impaired parenchymal development—further reinforcing that maintaining an adequate MP:ME ratio is crucial. Beyond parenchymal development, long-term performance implications are also linked to the MP:ME ratio. Excessive fat deposition in the mammary gland during the allometric growth phase (3–10 mo) is consistently associated with lower milk yield in the first lactation (Capuco et al., 1995; Sejrsen and Purup, 1997; Albino et al., 2017). Conversely, maintaining growth with an adequate MP supply allows heifers to reach breeding size earlier without impairing mammary development, potentially reducing age at first calving without compromising future milk yield (Silva et al., 2002; Albino et al., 2015; Weller et al., 2016). One nutritional strategy to improve the MP:ME ratio without excessively raising dietary CP is to increase the proportion of rumen-undegradable protein (RUP). Diets with higher RUP enhance MP supply because a larger fraction of protein escapes rumen degradation and contributes directly to intestinal amino acid absorption. Silva et al. (2018) showed that Holstein heifers fed diets with approximately 51% of CP as RUP achieved greater average daily gain, feed efficiency, and N retention compared with lower RUP levels, without compromising mammary ultrasonography traits. This aligns with earlier evidence that additional dietary protein supplied as RUP (e.g., from fish meal or heat- treated soybean meal) sustains growth while reducing the risk of excessive mammary fat deposition (Moallem et al., 2004). Thus, increasing RUP is a practical approach to elevate MP intake while keeping dietary CP moderate, improving efficiency and mitigating environmental N excretion. In summary, the MP:ME ratio is a key nutritional lever in postweaned heifer diets. Higher ratios favor mammary parenchymal development and reduce the likelihood of excess fat deposition, supporting both heifer growth efficiency and long-term milk production potential. Strategic use of high-RUP feeds can help meet MP requirements without resorting to overly high dietary CP, aligning with both performance and sustainability goals. Traditional nutrition models: Foundations and limitations Nutrition models for calf and heifer growth have historically relied on empirical and mechanistic functions that link BW to age and nutrient intake. The NRC (2001) and NASEM (2021) frameworks, for example, calculate nutrient requirements for energy and protein deposition based on BW, target ADG, and mature size. Growth itself is often represented using simple functions such as the Brody, Gompertz, or Richards equations (Richards, 1959; Owens et al., 1993). While useful, these models present clear limitations. First, they often treat growth as a unidimensional process, relying on BW as the sole indicator. This approach ignores the importance of skeletal growth and body composition, both of which are crucial for predicting developmental milestones like puberty or breeding readiness (Heinrichs & Losinger, 1998). Second, most validation datasets are derived from North American or European populations, which may not proper reflect genetic selection or management systems in tropical or subtropical environments (Silva et al., 2021). Finally, traditional models rarely incorporate variability across individuals, even though differences in tissue accretion rates or feed efficiency strongly influence outcomes. The result is that models provide general targets but often fall short when applied to specific herds, breeds, or climates. Finally, most frameworks do not account for the incidence of disease or its effects on nutrient requirements and tissue accretion, despite the profound impact of morbidity on calf growth trajectories. Figure 1. Relationship between predicted BW (kg; x-axis) from the CalfSim tool and observed BW (kg; y-axis) from 27 dairy calf nutrition studies extracted during the literature review. The dashed black line represents the line of equality (X = Y), and the solid red line represents the fitted regression line. Main model assessment statistics are displayed in the inset box within the scatter plot (upper left). The descriptive statistics of observed and predicted BW, including the minimum (Min.), first quartile (Q1), median, mean, third quartile (Q3), maximum (Max.), and SD, are shown in the bottom right. The histograms above and to the right of the graph show the distribution of predicted and observed values, respectively. RMSE = root mean square error; CCC = concordance correlation coefficient; σstudy = study-level standard deviation; σres = residual standard deviation. In fact, when comparing the chapter dedicated to dairy calves in NRC (2001) with NASEM (2021), a notable evolution can be observed. The more recent framework not only details protein and energy requirements but also addresses calf starter intake more explicitly, with equations developed for both temperate and subtropical climates. Moreover, NASEM (2021) incorporated extensive model testing, validating its predictions against 397 literature treatment means. Building on this foundation, Da Silva and Costa (2025) developed the CalfSim tool (go.uvm.edu/calfsim), a web-based decision-support platform that uses NASEM (2021) equations as its backbone. Their evaluation of predicted versus observed BW, based on 27 studies from the literature, demonstrated both accuracy and precision of the NASEM (2021) predictions (Figure 1). However, it is important to highlight that in their regression analysis of predicted versus observed BW, the effect of study was modeled as a random effect (σstudy = 8.99 kg), with an intraclass correlation coefficient (ICC) of 0.59. This indicates a strong study effect, which in practical terms may be viewed as a proxy for “farm effect,” encompassing unmeasured factors such as housing, environmental conditions, health management, disease incidence, and genetics, as pointed out earlier. These contextual differences suggest that although the model performs well overall, prediction errors can emerge depending on the specific production conditions. A key aspect, and often a bottleneck when comparing studies or farms, is starter intake. Although new equations were proposed in NASEM (2021), and other formulations are available in the literature (e.g., Quigley et al., 2021; Silva et al., 2019), starter consumption is highly sensitive to factors such as management, environment, and health status. Because starter is a critical nutrient supply (particularly energy), this variable becomes especially important in systems with lower milk allowances and after the first month of life. Accurate equations for predicting starter intake are therefore fundamental to obtaining growth predictions that more closely reflect on-farm reality. Looking ahead, one promising avenue lies in the development of hybrid models, which is a mathematical modeling paradigm that can aggregate empirical or semi- mechanistic nutrition models with machine learning approaches (Tedeschi, 2023). Such hybrid models could refine predictions by incorporating additional variables not currently considered in nutrition models, such as disease occurrence, behavioral indicators, or environmental stressors. However, these approaches require access to larger and richer datasets, which may be increasingly feasible through Precision Livestock Farming Technologies (PLF), such as wearable sensors, automated feeders, and continuous behavioral monitoring systems (Costa et al., 2021). The integration of nutritional models with data-driven corrections represents a promising strategy to enhance predictive capacity and support on-farm decision making in diverse production environments Considerations about age at first calving AFC is a pivotal outcome shaped by nutrition and growth, with profound implications for lifetime productivity, health, and culling risk. Traditional recommendations in the United States encourage calving between 22 and 24 months to minimize rearing costs and accelerate returns (Hoffman, 1997; Ettema & Santos, 2004). However, recent evidence highlights that the relationship between AFC, calving difficulty, and long-term milk production is more complex than previously assumed. In a large dataset covering over 687 herds and 1 million calving observations across the United States, Marcondes et al. (2025) show that heifers calving later, at around 27–28 months, produced more milk across their lifetime and likely benefited from additional time for mammary development. However, these older heifers also experienced higher levels of dystocia, which depressed first-lactation yield and increased health risks. This highlights the trade-off: while delayed AFC may enhance mammary development and lifetime output, it also increases the likelihood of calving complications. For this reason, the overall recommendation continues to favor targeting 23–24 months as the ideal AFC. At this age, heifers have reached sufficient skeletal maturity to minimize calving difficulty while still entering the herd early enough to balance rearing costs with productive life. The challenge today is that many farms are pushing AFC lower, to around 21–22 months, in an effort to reduce the high costs of heifer rearing, which often exceed $3,000 per animal. While this strategy reduces upfront expenses, the assumption that early calving has little to no effect on first-lactation performance is not supported by the evidence. The findings of Marcondes et al. (2025) demonstrate that heifers calving at 21–22 months not only produced less milk at first parturition but also faced a greater likelihood of dystocia, compounding the negative impact on both performance and welfare. Thus, the re-evaluation of nutrition models must account not just for growth rate and cost efficiency, but also for how growth trajectories interact with biological maturity, calving ease, and long-term productivity. The ultimate goal should be to support breeding strategies that consistently achieve 23–24 months AFC while avoiding the risks of both too-early and too-late calving. New tools and directions The advent of precision livestock farming (PLF) technologies offers new opportunities to improve growth monitoring and prediction in dairy cattle. Advances in automated imaging, computer vision, and 3-dimensional systems enable objective assessment of body weight, body condition, and conformation, reducing reliance on manual scoring and generating high-frequency data for longitudinal analysis (Azzaro et al., 2011; Salau et al., 2017; Van Hertem et al., 2020; Xavier et al., 2022). Walk-over weighing systems also provide continuous and labor-efficient body weight monitoring with strong agreement to conventional scales (Dickinson et al., 2013). Beyond measurement, predictive analytics are being applied to these data streams: machine-learning algorithms trained on image features can accurately predict body weight and track composition of body weight changes throughout lactation (Song et al., 2018; Xavier et al., 2022). Importantly, because most farmers do not routinely weigh their growing heifers or evaluate body condition score (BCS), camera-based systems offer a practical and impactful improvement in assessing heifer development. Looking ahead, these tools will not only enable body weight monitoring but also withers height measurement and analysis of the BW:height relationship. Such metrics can help determine whether accelerated growth programs (higher ADG targets) are driving true structural development (muscle and bone) rather than undesirable fattening. Combining these PLF tools with traditional nutrition models provides a path to refining growth predictions in a way that is both biologically meaningful and practically useful. Summary Despite advances, significant challenges remain. Growth trajectories vary across breeds, regions, and management systems, limiting the transferability of models. Longitudinal datasets spanning multiple lactations remain rare, making it difficult to validate how early-life growth influences lifetime productivity. Additionally, mechanistic models such as NASEM (2021) require extensive input data that may not be readily available on commercial farms, while simpler empirical curves may lack biological depth. The way forward lies in integrating biological understanding with modern data streams. Nutrition models must account for both weight and structural growth, incorporate early-life programming effects, and leverage precision technologies to reflect real-world variability. Re-evaluating models in this way will allow for more accurate feeding strategies, optimized AFC, and improved lifetime productivity. References Albino, R. L., A. L. Sguizzato, K. M. Daniels, M. S. Duarte, M. M. Lopes, S. E. F. Guimarães, M. M. D. C. A. Weller, and M. I. Marcondes. 2017. Performance strategies affect mammary gland development in prepubertal heifers. J. Dairy Sci. 100:8033–8042. https://doi.org/10.3168/jds.2016-12489 Albino, R. L., M. I. Marcondes, R. M. Akers, E. Detmann, B. C. Carvalho, and T. E. Silva. 2015. Mammary gland development of dairy heifers fed diets containing increasing levels of metabolisable protein: metabolisable energy. J. Dairy Res. 82:113–120. https://doi.org/10.1017/S0022029914000697 Azzaro, G., G. Caccamo, S. Ferguson, G. Battiato, A. Farinella, G. Puglisi, D. Petriglieri, and A. Licitra. 2011. Objective estimation of body condition score by a 3- dimensional automatic system in dairy cows. J. Dairy Sci. 94:2126– 2137.https://doi.org/10.3168/jds.2010-3467. Capuco, A. V., J. J. Smith, D. R. Waldo, and C. E. Rexroad Jr. 1995. Influence of prepubertal dietary regimen on mammary growth of Holstein heifers. J. Dairy Sci. 78:2709–2725. https://doi.org/10.3168/jds.S0022-0302(95)76902-8 Costa, J. H. C., Cantor, M. C., Neave, H. W. 2021. Symposium review: Precision technologies for dairy calves and management applications. J. Dairy Sci., 104(1):1203-1219. https://doi.org/10.3168/jds.2019-17885 D. L. Van De Stroet, J. A. Calderón Díaz, K. J. Stalder, A. J. Heinrichs, C. D. Dechow. Association of calf growth traits with production characteristics in dairy cattle. Journal of Dairy Science. 2016 Oct;99(10):8347-8355. DOI: 10.3168/jds.2015- 10738. Da Silva, T. E., Costa, J. H. C. 2025. CalfSim tool: A free and user-friendly decision support tool for designing and simulating optimized feeding plans for dairy calves— A prediction assessment study. J. Dairy Sci. Comm. 5:654-659. https://doi.org/10.3168/jdsc.2025-0777 Davis Rincker, L. E., VandeHaar, M. J., Wolf, C. A., Liesman, J. S., Chapin, L. T., & Weber Nielsen, M. S. (2011). Effect of intensified feeding of heifer calves on growth, pubertal age, calving age, milk yield, and economics. Journal of Dairy Science, 94(7), 3554–3567.DOI: 10.3168/jds.2010-3923 Dickinson, R. A., N. A. Lopez-Villalobos, J. A. Jago, K. L. Matthews, S. A. Correa-Luna, and R. A. Hickson. 2013. An automated walk-over weighing system as a tool for measuring liveweight change in lactating dairy cows. J. Dairy Sci. 96:4477–4486. https://doi.org/10.3168/jds.2012-6522 Ettema, J. F., & Santos, J. E. P. (2004). Impact of age at calving on lactation, reproduction, health, and income in first-parity Holsteins on commercial farms. Journal of Dairy Science, 87(8), 2730–2742.https://doi.org/10.3168/jds.S0022- 0302(04)73400-1 Heinrichs, A. J. (1993). Raising dairy replacements to meet the needs of the 21st century. Journal of Dairy Science, 76(10), 3179–3187.https://doi.org/10.3168/jds.S0022- 0302(93)77656-0 Heinrichs, A. J., & Losinger, W. C. (1998). Growth of Holstein dairy heifers in the United States. Journal of Animal Science, 76(5), 1254–1260. Hoffman, P. C. (1997). Optimum body size of Holstein replacement heifers. Journal of Animal Science, 75(3), 836–845.DOI: 10.2527/1997.753836x Kusaka H., Yamazaki T., Sakaguchi, M. Association of age at first calving with longevity, milk yield, and fertility up to the third lactation in a herd of Holstein dairy cows in Japan. Journal of Reproduction and Development. Vol. 69, No. 6 (2023), pp. 291- 297.https://doi.org/10.1262/jrd.2023-012. Lanier, J.S., F. K. McKeith, N. A. Janovick, R. A. Molano, M. E. Van Amburgh, and J. K. Drackley. 2021. Influence of starter crude protein content on growth and body composition of dairy calves in an enhanced early nutrition program. J. Dairy Sci. 104:3082-3097. doi:10.3168/jds.2020-19580. Marcondes, M. I., and Silva, A. L. 2021. Determination of energy and protein requirements of preweaned dairy calves: A multistudy approach. J. Dairy Sci. 104:10536–10554. https://doi.org/10.3168/jds.2021-20272 Marcondes, M. I., Chagas, J. C. C., Kertz, A. F., Allen, T., & Bewley, J. M. (2025). Connecting the dots—Calving difficulty, age at first calving, and enhanced cow production. Applied Animal Science, 41(3), 237– 251.https://doi.org/10.15232/aas.2024-02620 Moallem, U., G. E. Dahl, E. K. Duffey, A. V. Capuco, D. L. Wood, K. R. McLeod, R. L. Baldwin VI, and R. A. Erdman. 2004. Bovine somatotropin and rumen- undegradable protein effects in prepubertal dairy heifers: effects on body composition and organ and tissue weights. J. Dairy Sci. 87:3869-3880. https://doi.org/10.3168/jds.S0022-0302(04)73526-2 NASEM (National Academies of Sciences, Engineering, and Medicine). 2021. Nutrient Requirements of Dairy Cattle, 8th rev. ed. National Academies Press, Washington, DC. NRC. 2001. Nutrient Requirements of Dairy Cattle. 7th rev. ed. Natl. Acad. Sci., Washington, DC. Owens, F. N., Dubeski, P., & Hanson, C. F. (1993). Factors that alter the growth and development of ruminants. Journal of Animal Science, 71(11), 3138–3150. Pinheiro, J. S., P. S. D. Silva, de Andrade, D. R., E. F. Santos, A. C. Costa, N. V. Trópia, S. C. Valadares Filho, E. Detmann, S. E. F. Guimarães, and M. I. Marcondes. 2025. Influence of different amounts of milk replacer on esophageal leakage, rumen fermentation characteristics, gastrointestinal tract passage rate, and microbial crude protein synthesis of nursling animals. J. Dairy Sci. 108:5816–5831. https://doi.org/10.3168/jds.2024-25695 Quigley, J. D., Dennis, T. S., Suarez-Mena, F. X., Chapman, C. E., Hill, T. M., Aragona, K. M. 2021. Models to predict dry feed intake in Holstein calves to 4 months of age. J. Dairy Sci. 104:5539-5556. https://doi.org/10.3168/jds.2020-19581 Richards, F. J. (1959). A flexible growth function for empirical use. Journal of Experimental Botany, 10(2), 290–301. Salau, J., J. H. Haas, W. Junge, and G. Thaller. 2017. A multi-Kinect cow scanning system: Calculating linear traits from manually marked recordings of Holstein– Friesian dairy cows. Biosyst. Eng. 157:92– 98.https://doi.org/10.1016/j.biosystemseng.2016.10.008. Sejrsen, K., and S. Purup. 1997. Influence of prepubertal feeding level on milk yield potential of dairy heifers: a review. J. Anim. Sci. 75:828–835.DOI: 10.2527/1997.753828x Sejrsen, K., Huber, J. T., Tucker, H. A., & Akers, R. M. (1982). Influence of nutrition on mammary development in pre- and postpubertal heifers. Journal of Dairy Science, 65(5), 793–800.DOI: 10.3168/jds.s0022-0302(82)82268-6 Silva, A. C. R. T., A. L. L. Sguizzato, M. M. Campos, P. P. Rotta, M. I. Marcondes, J. R. R. Dorea, G. L. Menezes, J. V. F. Vieira, and A. L. Silva. 2025. Impact of reducing starter feed crude protein content with different milk allowances on the performance and nitrogen balance in dairy calves. J. Dairy Sci. In press. Silva, A. L., DeVries, T. J., Fernandes, E. C., & Marcondes, M. I. (2021). Development and evaluation of equations to predict growth of Holstein dairy heifers in a tropical climate. Journal of Dairy Science, 104(1), 102–112. Silva, A. L., DeVries, T. J., Tedeschi, L. O., Marcondes, M. I. 2019. Development of equations, based on milk intake, to predict starter feed intake of preweaned dairy calves. animal, 13:83-89. https://doi.org/10.1017/S1751731118000666 Silva, A. L., E. Detmann, J. Dijkstra, A. M. Pedroso, L. H. P. Silva, A. F. Machado, F. C. Sousa, G. B. dos Santos, and M. I. Marcondes. 2018. Effects of rumen- undegradable protein on intake, performance, and mammary gland development in prepubertal and pubertal dairy heifers. J. Dairy Sci. 101:5991–6001. https://doi.org/10.3168/jds.2017-13230 Silva, L. F. P., M. J. VandeHaar, B. K. Whitlock, R. P. Radcliff, and H. A. Tucker. 2002. Relationship between body growth and mammary development in dairy heifers. J. Dairy Sci. 85:2600–2602. https://doi.org/10.3168/jds.S0022-0302(02)74344-0 Soberon, F., Raffrenato, E., Everett, R. W., & Van Amburgh, M. E. (2012). Preweaning milk replacer intake and effects on long-term productivity of dairy calves. Journal of Dairy Science, 95(2), 783–793. DOI: 10.3168/jds.2011-4391 Song, X., C. Bokkers, P. van Mourik, P. Groot Koerkamp, and T. van der Tol. 2018. Automated body weight prediction of dairy cows using 3-dimensional vision. J. Dairy Sci. 101:4447–4459. https://doi.org/10.3168/jds.2017-13094 Stamey, J. A., Janovick, N. A., Kertz, A. F., and Drackley, J. K. 2012. Influence of starter protein content on growth of dairy calves in an enhanced early nutrition program. J. Dairy Sci. 95:3327–3336. DOI: 10.3168/jds.2011-5107 Tedeschi, L. O. 2023. Review: The prevailing mathematical modeling classifications and paradigms to support the advancement of sustainable animal production. animal, 17(5):100813. https://doi.org/10.1016/j.animal.2023.100813 Van Amburgh, M. E., D. M. Galton, D. E. Bauman, R. W. Everett, D. G. Fox, L. E. Chase, and H. N. Erb. “Effects of three prepubertal body growth rates on performance of Holstein heifers during first lactation.” Journal of Dairy Science, Vol. 81, No. 2 (1998), pp. 527-538. DOI: 10.3168/jds.S0022-0302(98)75604-8 Van Hertem, T., I. T. Pastell, L. Viazzi, C. Bahr, A. Schlageter-Tello, D. Berckmans, and E. M. Tullo. 2020. Implementation of an automated 3-dimensional vision system for dairy cow body trait estimation under field conditions. Comput. Electron. Agric. 168:105140. https://doi.org/10.1016/j.biosystemseng.2017.08.011. Weller, M. M. D. C. A., R. L. Albino, M. I. Marcondes, W. Silva, K. M. Daniels, M. M. Campos, M. S. Duarte, M. L. Mescouto, F. F. Silva, and S. E. F. Guimarães. 2016. Effects of nutrient intake level on mammary parenchyma growth and gene expression in crossbred (Holstein × Gyr) prepubertal heifers. J. Dairy Sci. 99:9962–9973. https://doi.org/10.3168/jds.2016-11532 Xavier, C., S. C. Valente, F. Nigsch, A. Fischer, and L. Hulsegge. 2022. The use of 3- dimensional imaging of Holstein cows to estimate body weight and monitor the composition of body weight change throughout lactation. J. Dairy Sci. 105:4508- 4519. https://doi.org/10.3168/jds.2021-21337. Yousefinejad, S., F. Fattahnia, M. Kazemi-Bonchenari, H. Khanaki, J. K. Drackley, and M. H. Ghaffari. 2021. Soybean oil supplementation and starter protein content: Effects on growth performance, digestibility, ruminal fermentation, and urinary purine derivatives of Holstein dairy calves. J. Dairy Sci. 104:1630-1644. doi:10.3168/jds.2020-18823.