Effect of weaning age on nursery pig and sow reproductive performance

Efecto de la edad al destete en el desempeño de los lechones en el destete y reproductivo de hembras

Effet de l’âge du sevrage sur les porcs en pouponnière et les performances de reproduction des truies

Alison L. Smith, MSc; Kenneth J. Stalder, MSc, PhD; Timo V. Serenius, MSc, PhD; Tom J. Baas, MSc, PhD; John W. Mabry, MSc, PhD

Department of Animal Science, Iowa State University, Ames, Iowa. Corresponding author: Dr Kenneth J. Stalder, Department of Animal Science, Iowa State University, 109 Kildee Hall, Ames, IA 50011-3150; Tel: 515-294-4683; Fax: 515-294-5698; E-mail: stalder@iastate.edu.

Cite as: Smith AL, Stalder KJ, Serenius TV, et al. Effect of weaning age on nursery pig and sow reproductive performance. J Swine Health Prod. 2008;16(3):131–137..
Also available as a PDF.

Keywords: swine, weaning age, sows, wean-to-first-service interval
Search the AASV web site for pages with similar keywords.

Received: March 8, 2007
Accepted: November 12, 2007

The swine industry has shifted to earlier weaning to improve farrowing-crate utilization, increase numbers of pigs born per sow per year, improve piglet health, and increase the number of pigs produced at a facility in a year.¹ Many producers and scientists are re-evaluating weaning-age decisions, comparing growth differences and herd-health issues among pigs weaned at different ages. Lactation length (weaning age) can impact both nursery growth and sow fertility and should be optimized so that producers can maximize profitability of their pork operations.

Segregated early weaning (SEW) was developed to minimize transmission of pathogens from sow to piglet. This process consists of farrowing sows on the same site as the rest of the breeding herd, weaning piglets at 10 to 21 days of age, and as a result decreasing the amount of medication needed.¹ Applying these management practices improves piglet health, but also impacts piglet growth and sow reproductive performance. Even though most of the benefits of weaning at a later age are observed during the early postweaning period, the effect has been reported to persist through the grow-finish phase.²

Sow reproductive performance is dependent upon various factors, eg, sow breed, parity, and environment, but management decisions such as lactation length also influence fertility. Short lactations may negatively affect measures such as weaning-to-first-service interval (WSI).³ However, others have reported that lactation length has no effect on WSI and subsequent litter size.⁴ Sow performance differences observed are likely the result of the range in lactation length.

Limited research has been conducted to determine the weaning age that results in optimum performance of the pig in the early stages of its development and in reproductive performance of the sow. The objectives of this research were to determine the effect of weaning age on nursery-pig performance and sow reproductive performance.

Materials and methods

Study animals

Data was collected over a period of 1 year in a 600-sow farrow-to-finish operation farrowing approximately 25 litters per week. The study included Danbred N.A. (Columbus, Nebraska) Yorkshire × Landrace barrows and gilts (n = 2467) from a commercial maternal-line multiplication herd (H&K Enterprises, Nevada, Iowa). All pigs were produced by purebred Landrace sows in their first to eleventh parities, divided into six parity classes for the purposes of this study. Parities > 5 were combined into a single parity class.

Study design

Each piglet was identified and weighed and sex was determined within 24 hours of birth. Within 3 days post farrowing, litter size was standardized to 10 to 11 piglets per litter by cross-fostering individuals within treatment. By 7 days of age, all boars were castrated.

Each litter was assigned at birth to one of two weaning-age or lactation-length treatments, ie, the same groups were described differently for application to different populations of animals (pigs and sows). The first weaning-age (lactation-length) treatment, average 15 days of age at weaning, included pigs that were weaned at 14, 15, and 16 days of age. The second treatment, average 20 days of age at weaning, included pigs weaned at 19, 20, and 21 days of age. A weighted average weaning age, based on the number of pigs weaned at each age, was also calculated. Weighted average weaning-age (lactation-length) treatments were 14.9 days of age and 20.2 days of age.

Litters were weaned once a week, at which time piglets in each treatment group were weighed and randomly assigned by treatment to a nursery pen. Pigs remained in their pens until 42 days post weaning, when they were weighed and removed from test (off-test weight). Seven or eight pens of pigs (three or four pens per treatment) were evaluated per replication. A total of 12 replications were completed, including a total of 89 pens of pigs evaluated (45 and 44 pens of average 15 days and average 20 days at weaning, respectively).

The experimental protocol followed the operation’s production practices and met or exceeded requirements in Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching.⁵

Exclusion criteria

Individual pigs that were substandard or non-ambulatory at weaning were not included in the nursery portion of the trial. A substandard pig was defined as a pig falling outside a target market-weight range and substantially discounted from an established base price or value.⁶ Pigs were removed from test pens prior to 42 days post weaning due to mortality or if a health condition existed in which the pig did not respond to medical treatment. Morbidity was defined as pigs appearing ill, diseased, unthrifty, or characterized by loss of body weight.⁷

Housing and feeding

Pigs were housed at 26 to 28 per pen (n = 2467). Each nursery room contained eight pens, four for each weaning-age treatment, and nursery pens (2.44 × 3.05 m) provided 0.27 to 0.29 m² per pig. Each pen was equipped with a single-sided stainless steel self-feeder (Chore-Time, Milford, Indiana; 76.2 cm linear trough space per pen, 2.72 to 2.93 cm per pig) and two nipple drinkers, providing ad libitum access to feed and water. In both weaning-age groups, pigs were fed a four-phase diet regime from weaning to 42 days post weaning. A feed budget was developed in which each pig was provided 1.25 kg of a 1.70% lysine pelleted diet, 6.14 kg of a 1.50% lysine meal diet, and 12.57 kg of a 1.30% lysine meal diet, with the remainder a 1.20% lysine meal diet (as-fed basis). Feed disappearance was recorded on a pen basis every day of the trial.

Data management

Nursery performance data was collected and entered into Microsoft Excel (1999; Microsoft Corporation, Redmond, Washington). Growth and feed efficiency were calculated using body weight at weaning and 42 days post weaning.

For sow reproductive performance, breeding, farrowing, and weaning information was recorded in PigWIN (Version 5, 2005; Decision Power Inc, Little Canada, Minnesota), and relevant data was extracted from that program for analysis.

Sow management

A total of 339 purebred Landrace sows were included in this experiment (Table 1). Sows were mated using pooled-sire semen resulting in Yorkshire × Landrace crossbred pigs. At weaning, sows were housed in breeding stalls (55.9 cm × 182.9 cm) in mechanically ventilated buildings. Each breeding stall was equipped with a water trough that was filled twice daily, providing ad libitum access to water. While in the breeding stall, each sow was fed 2.7 kg per day of a 14.4% crude protein, 0.7% lysine diet (as-fed basis) (Table 2).

Estrus detection began 4 days post weaning and was performed once daily at approximately 7:30 am by providing stall-line contact with a mature boar. Estrus detection continued until each sow was mated or the decision to cull an individual sow was made. Sows were mated by artificial insemination 24 and 48 hours post estrus detection.

Approximately 2 days post mating, sows were moved from the breeding stalls into pens with partially slatted concrete floors in mechanically ventilated facilities. Bred sows were grouped by size and penned with five to nine other females, with 1.26 m² to 1.30 m² per sow. Each pen was equipped with a nipple drinker that provided ad libitum access to water. Sows were fed 2.7 kg per day of the gestation diet described (Table 2).

Pregnancy detection was performed 28 days post first mating using real-time ultrasound (Alliance Medical Inc, Montreal, Quebec, Canada). Sows detected not pregnant were moved back to the breeding area for estrus detection and mating. Pregnant sows were housed in the same pens until approximately 5 days before farrowing, when they were moved to farrowing facilities.

Sows farrowed in mechanically ventilated buildings in farrowing stalls (55.9 cm × 182.9 cm) that were each equipped with a nipple drinker providing ad libitum access to water. Lactating sows were fed a diet containing 18.3% crude protein and 1% lysine (as-fed basis; Table 2) three times daily in a step-up program that started with 0.9 kg per feeding immediately after farrowing and was slowly increased as the sow fully consumed the feed provided, to a maximum intake of approximately 8.2 kg per day.

Calculations and statistical analysis

Data were analyzed by analysis of variance using PROC MIXED of SAS (SAS, 2003, Cary, North Carolina). Variables included number of pigs, weaning weight, average daily gain (ADG), average daily feed intake (ADFI), gain:feed ratio (G:F), 42-day weight, feed costs per kg of gain, wean-to-service interval (WSI), total number of piglets born, and total number born alive. Variables and residuals were tested for normality. When fixed effects were a significant source of variation, differences were determined using the PDIFF option of SAS. A chi-square test for proportions was utilized to evaluate mortality and morbidity differences between the two weaning-age treatment groups. The number of days from weaning until the sow conceived her next litter (weaning-to-conception interval; WCI) was used as a measure of the sow’s reproductive performance. If the sow conceived on the first breeding after weaning, the WSI and WCI were considered the same (PigWIN). If the sow failed to conceive on the first postweaning estrus (repeat breeder) and was bred again on the next estrus and became pregnant, WCI was counted from the day of weaning to the second breeding.

A chi-square test for proportions was utilized to evaluate the number of sows removed from the herd versus those retained in the herd and the number of sows with normal WSI (≤ 10 days)⁸ versus those with abnormal WSI (> 10 days), comparing the two lactation-length treatments.

Immediately after completion of the trial, pig data were subjected to a distribution evaluation. Records with off-test weights > 33.6 kg (one pig) or < 5.9 kg (26 pigs) or ADG > 1.54 kg day (14 pigs) or < -0.08 kg per day (15 pigs) were excluded from the analysis, as these outliers were likely related to management factors (for example, failure to remove pigs falling behind or to identify sick animals in a timely manner) or errors in measurement. Records were deleted by simply removing from analysis the animals having extreme outlying data. This resulted in a total of 2411 pigs included in the statistical analysis.

Pig was the experimental unit for all variables measured in the individual animal. Pen was the experimental unit for G:F, ADFI, and feed cost per kg gain. Fixed effects of weaning-age treatment, parity of dam, and pen within weaning-age treatment were included in the model for ADG, and birth weight was included as a linear covariate. Sow ID and group ID were initially included as random effects; however, they had little impact (extremely low variance) and were excluded from the analysis model for ADG. The two-way interactions between fixed effects were tested but were dropped from the models because they were not a significant source of variation. Fixed effects of weaning-age treatment and pen within weaning-age group were included in the models for ADFI and G:F.

Sows with records for WSI and WCI were removed according to the following criteria: WSI > 50 days (four sows) and WCI > 100 days (eight sows). A binary response variable was created for evaluation of culling rate: removed from the herd (0), or retained in the herd (1). Culled sows were included in the analysis only if they were culled for reproductive failure for reasons that were likely the result of a treatment effect. Sows culled because of management decisions or other factors not related to the treatments were excluded from the analysis. Weaning-to-first-service interval was also modeled using a binary response variable: WSI >10 days (2) or WSI ≤ 10 days (1).

Fixed effects of weaning month, lactation-length treatment, and parity of the sow were included in the analysis of WSI, WCI, total born, and total born alive. A covariate for WSI was included in the model used to analyze WCI. The two-way interactions between fixed effects were tested but were dropped from the models because they were not significant sources of variation for the traits evaluated.

Results

Nursery-pig performance

In the group weaned at average 20 days of age, pigs were 1.53 kg heavier at weaning, nursery ADG was greater by 0.08 kg per day, and ADFI was greater by 0.07 kg per day than in pigs weaned at average 15 days of age (Table 3). A linear birth-weight covariate (P < .01) stated that for every 1-kg increase in birth weight there was a 0.18-kg increase in ADG. Off-test weight was greater in the group weaned at average 20 days of age (Table 3).

Morbidity in pigs weaned at average 20 days of age was lower than in pigs weaned at average 15 days of age (Table 3). Mortality rates did not differ between the weaning-age treatments (Table 3).

Gain-to-feed ratio was similar for pigs in the two weaning-age treatments, but feed cost per kg gain was greater by $0.03 in the average 15 days weaning-age group than in the average 20 days group (Table 3). This reflects the higher ADG and lower morbidity observed in the pigs weaned at an average of 20 days of age.

Sow performance

A total of 239 sows were included in the statistical analysis. Sixty sows were culled for the following reasons: 31 for reproductive failure, 17 for body condition, seven for old age, and five for miscellaneous reasons. An additional 40 sows that farrowed during the weaning-age treatment either died or were not included in the analysis due to reasons unrelated to the weaning-age treatment.

No significant lactation-length treatment differences between the 15-day and 20-day lactation lengths were observed for WSI (7.98 versus 8.38 days, SD 0.78), WCI (15.65 versus 16.93 days, SD 2.56), total born (11.15 versus 10.97, SD 0.41), total born alive (9.28 versus 9.41, SD 0.56), and culling rate (6.06% versus 7.95%) (P > .05). A linear WSI covariate (P < .01) stated that for every 1-day increase in WSI there was a 0.7-day increase in WCI. No differences were observed (P > .05) between the treatment groups in the numbers of sows with normal WSI (≤ 10 days) and abnormal WSI (> 10 days). Sows in the average 15 days group included 98 with normal and 23 with abnormal WSI. Sows in the average 20 days group included 100 with normal and 18 with abnormal WSI. However, the coefficient of variation for WSI decreased as lactation length increased when sows with normal WSI (198 sows) and abnormal WSI (41 sows) were compared (data not shown).

Parity differences for WSI, total born, and total born alive were observed (Table 4). Mean WSI was greater in Parity 2 sows (P < .01) than parities higher than 2. Total born and total born alive were greater in Parity 2 and Parity 4 sows than in sows of parity > 5 (P < .01; Table 4). Total born alive was greater in Parity 2 and 4 sows than in Parity 3 and 6 sows (P < .01). As parity increased, WSI decreased, and there was a linear decrease in total born alive in parities ≥ 4. There was no difference in culling rate for reproductive failure between sows weaned at average 15 days (6.06%) and those weaned at average 20 days (7.95%) (P = .43).

Discussion

The results of this study agree with those of Wolter and Ellis⁹ and Himmelberg et al,¹⁰ who reported a favorable correlation between weaning weight and weaning age. Wolter and Ellis⁹ studied pigs from the Pig Improvement Company (PIC) and reported that pigs heavier at weaning were heavier at birth and at 56 days of age and reached slaughter weight 8.6 days sooner. Mahan and Lepine¹¹ and Roberts¹² also reported that weaning weight was predicative of overall pig performance and days to market weight in crossbred pigs. Studies^2,10 have shown that feed intake and rate of gain during the nursery period increase with weight at weaning. Main et al² also reported highly favorable correlations between weaning age, ADG, and weight sold per pig weaned in a multi-site system utilizing crossbred pigs.

Pigs weaned in the average 20 days group were 3.13 kg heavier than pigs weaned in the average 15 days group. The heavier weights obtained by the average 20 days weaning-age group might have been influenced by the fact that both treatment groups were fed alike, receiving budgeted feed amounts geared to meet the needs of the average 15 days weaning-age group. Main et al,² studying pigs from a commercial PIC operation, also observed a significantly higher nursery off-test weight (greater by 5.5 kg) in pigs weaned at 20 days of age than in those weaned at 15 days of age. The results of the current study reported on weights only through 42 days post weaning, because the production system where the trial was conducted typically sold many of the pigs at this point and further data collection was not possible. Main et al² and Powell and Aberle¹³ reported that the higher ADG and lower mortality that were a result of weaning at an older age largely occurred in the initial 42 days after weaning, with some further improvements in ADG and lower mortality through finishing.

Results for ADG and ADFI are consistent with those of Himmelberg et al¹⁰ and Leibbrandt et al,¹⁴ who reported that ADG and ADFI were significantly greater throughout the nursery and grow-finish phases of production in pigs that were heavier at weaning than in their lighter littermates. Hohenshell et al¹⁵ also observed lower post-weaning weight gains associated with early weaning. Additionally, Main et al² and Fangman et al¹⁶ reported greater nursery ADG and ADFI as weaning age increased from 12 days to 21 days.

Weaning age is an important driver of nursery costs, but mortality and morbidity within weaning age are important factors in determining the optimal weaning age for individual pork operations. However, in this study, mortality and morbidity in both treatments would be considered above average by current industry standards.¹⁷ These results are consistent with those reported by Main et al,² who found that mortality was 2.82% in pigs weaned at 15 days of age and 0.54% in pigs weaned at 20 days of age. The greater mortality observed in pigs weaned at younger ages represents a substantial reduction of net income through lost revenue from pigs that died and increased expenses of feeding the pigs until they died. In a study by Deen,¹⁸ variability in profits due to cull, dead, and lightweight pigs was > 50%, while the variability in ADG was > 15% and the variability in G:F was > 30%. This variability in profitability of cull, dead, and lightweight pigs reveals the importance of minimizing numbers of these pigs in any production system. Mortality and morbidity in the nursery are largely a function of entry and exit weights, and weaning age significantly impacts both of these weights.

Mahan et al¹⁹ also reported similar feed conversion ratios for pigs with different weaning weights. The current study showed that weaning age and weaning weight are related. Main et al² reported that G:F ratios were lower in pigs weaned at 12 days than in pigs weaned at 21 days. Additionally, the same study found that G:F ratios were similar among pigs weaned at 15 days, 18 days, and 21 days. Schinckel and de Lange²⁰ portray the relationships between pig genotype and environmental factors. These relationships are essential in order to evaluate and implement different management strategies such as weaning age. The relationship between feed efficiency and body-weight gain was similar in the two weaning-age groups in the present study because their growth curves during the nursery phase of production were essentially the same.

Main et al²¹ reported that income over costs per pig weaned increased from $3.71 for pigs weaned at 12 days of age to $10.28 for pigs weaned at 21 days of age. Feed cost per kg of gain was not directly measured in this study. The current results demonstrate the importance of ADG, morbidity, and weaning age to profitability in the nursery phase of production in a pork operation.

Parities > 5 were combined in this study for two specific reasons. Subsequent reproductive performance for parties 6, 7, 8, 9, 10, and 11 was similar, and combining parities provided a more normal distribution of sows in each parity class across lactation-length treatment.⁶ Including the parity classes in the study provided results consistent with those reported by Tantasuparuk et al,⁴ who also utilized purebred females and evaluated a similar range of lactation lengths (17 to 24 days), compared to a range of 14 to 21 days in this study.

Previous studies have reported that increasing lactation length decreases WSI and increases subsequent litter sizes.^3,22,23 These studies analyzed sow records retrospectively, whereas in this study, determining effects of lactation length on sow productivity was an objective of the trial. The results of the previous studies^3,22,23 differ from those of the present investigation, likely due to differences in lactation length and the use of purebred sows versus crossbred sows in some cases. The results of the present study agree with those of Le Cozler et al,³ who reported that sow WSI decreased with increasing lactation length. However, in contrast to the present study, previous studies^4,23 reported that longer lactation lengths were associated with larger subsequent litter size. This difference is likely due to the relatively small range of the lactation-length treatments. Breed is an important source of variation in litter size, as purebred sows generally have lower prolificacy³ than crossbred females.²⁴ Since purebred Landrace females were utilized in this study, values for some reproductive traits may be lower than expected for crossbred females.

In a study by Xue et al,²⁵ parity affected WSI, total born, and total born alive, and average litter size was lower for parity 2 sows than for later parities. The relatively low number of observations in the parity subclasses in each weaning-age treatment in the present study might explain differences between the present and previous investigations. Thus, when analyzing the association between parity and the named traits, it is important to look at the correlation between WSI, WCI, and litter size by parity.

Previous studies have reported that sow longevity is adversely affected by reducing lactation length.^26,27 Xue et al²⁷ reported that lactation length was shorter (< 15 days) in sows removed from the herd than in sows that remained in the herd (> 16 days). Sows culled for reasons unrelated to reproduction were included in the Xue et al²⁷ study. Results may have been biased because culling might have been the result of factors other than lactation length.

The present study evaluated the difference between two average lactation lengths, 15 and 20 days, that are commonly implemented in commercial swine production systems today.¹ Mabray et al²⁸ reported that a lactation length < 14 days had a negative impact on subsequent performance and caused more variability in sow performance. Lactation lengths < 14 days were not included in the present study due to the possibility of a negative impact on subsequent sow performance. Other operations using different genetic lines or management systems may find different results and should develop their own qualifications for measuring subsequent sow reproductive performance.

Implications

• Weaning pigs at ≥ 20 days of age may prove advantageous in commercial operations because of improvements in nursery growth performance.

• Under the conditions of this study, feed costs in the first 42 days post weaning are lower when pigs are weaned at an average age of 20 days than at an average of 15 days.

• Weaning ages in the range of 14 to 21 days may not adversely impact sow reproductive performance.

• Parity of dam influences both piglet weights in subsequent phases of production and subsequent reproductive performance of the sow.

• The effects of different weaning ages should be assessed in the context of each operation.

Acknowledgements

This journal paper of the Iowa Agricultural and Home Economics Experiment Station, Ames, Iowa, Project No. 3600, was supported by Hatch Act and State of Iowa funds. The authors gratefully acknowledge H&K Enterprises, Inc, Nevada, Iowa, for allowing the study to be conducted at their facility and for technical assistance.

References

1. Harris DL. Multi-site Pig Production. Ames, Iowa: Iowa State University Press; 2000.

2. Main RG, Dritz SS, Tokach MD, Goodband RD, Nelssen JL. Increasing weaning age improves pig performance in a multi-site production system. J Anim Sci. 2004;82:1499–1507.

3. Le Cozler YJ, Dagorn J, Dourmad Y, Johansen S, Aumaître A. Effect of weaning-to-conception interval and lactation length on subsequent litter size in sows. Livest Prod Sci. 1997;51:1–11.

4. Tantasuparuk W, Lundeheim N, Dalin A-M, Kunavongkrit A, Einarsson S. Effects of lactation length and weaning-to-service interval on subsequent farrowing rate and litter size in Landrace and Yorkshire sows in Thailand. Theriogenology. 2000;54:1525–1538.

5. Federation of Animal Science Societies. Guide for the Care and Use of Agricultural Animals in Agricultural Research and Teaching. 1^st rev ed. Savoy, Illinois: Federation of Animal Science Societies; 1999.

*6. Baas TJ, Moeller S, Stalder KJ, Harmon J. Segregated Early Weaning. Distance Learning CD. Des Moines, Iowa: National Pork Board; 2004.

7. Corrigan BP. The Effects of Feeding Management on Growth Performance and Survivability of Newly Weaned Pigs [master’s thesis]. Urbana, Illinois: University of Illinois at Urbana-Champaign; 2002.

*8. PigCHAMP. PigCHAMP 2004 Datashare. PigCHAMP Inc; Ames, Iowa. 2004. Available at: www.pigchamp.com/2004YearendDatashare.htm. Accessed 19 February 2008.

9. Wolter BF, Ellis M. The effects of weaning weight and rate of growth immediately after weaning on subsequent pig growth performance and carcass characteristics. Can J Anim Sci. 2001;81:363–369.

10. Himmelberg LV, Peo ER Jr, Lewis AJ, Crenshaw JD. Weaning weight response of pigs to simple and complex diets. J Anim Sci. 1985;61:18–26.

11. Mahan DC, Lepine AJ. Effect of pig weaning weight and associated nursery feeding programs on subsequent performance to 105 kilograms body weight. J Anim Sci. 1991;69:1370–1378.

*12. Roberts J. Commercial nursery mortality and weaning weight. Proc Allen D. Leman Swine Conf. 2000;194–195.

13. Powell SE, Aberle ED. Effects of birth weight on growth and carcass composition of swine. J Anim Sci. 1980;50:860–868.

14. Leibbrandt VD, Ewan RC, Speer VC, Zimmerman DR. Effect of weaning and age at weaning on baby pig performance. J Anim Sci. 1975;40:1077–1080.

15. Hohenshell LM, Cunnick JE, Ford SP, Kattesh HG, Zimmerman DR, Wilson ME, Matteri RL, Carroll JA, Lay DC Jr. Few differences found between early- and late-weaned pigs raised in the same environment. J Anim Sci. 2000;78:38–49.

16. Fangman TJ, Tubbs RC, Becker BA, Allee GL, Misfeldt ML, Henningsen-Dyer K. Evaluation of segregated early weaning investigating performance, immunologic indicators (CD4, CD8), and herd health status. Swine Health Prod. 1996;4:217–229.

17. United States Department of Agriculture, Animal and Plant Health Inspection Service, Veterinary Services. NAHMS Swine 2000 Part III: Reference of Swine Health & Environmental Management in the United States, 2000. Available at: http://www.aphis.usda.gov/vs/ceah/ncahs/nahms/swine/swine2000/swine2kPt3.pdf. Accessed 2 March 2008.

*18. Deen J. Pro forma budgets and the at-risk pig. Proc Allen D. Leman Swine Conf. 2005;441–443.

19. Mahan DC, Cromwell GL, Ewan RC, Hamilton CR, Yen JT. Evaluation of the feeding duration of a phase 1 nursery diet to three-week-old pigs of two weaning weights. J Anim Sci. 1998;76:578–583.

20. Schinckel AP, de Lange CFM. Characterization of growth parameters needed as inputs for pig growth models. J Anim Sci. 1996;74:2021–2036.

21. Main RG, Dritz SS, Goodband RD, Tokach MD, Nelssen JL. Effects of weaning age on growing-pig costs and revenue in a multi-site production system. Swine Health Prod. 2005;13:189–197.

22. Tummaruk P, Luneheim N, Einarsson S, Dalin A-M. Reproductive performance of purebred Swedish Landrace and Swedish Yorkshire sows: II. Effect of mating type, weaning-to-first-service interval and lactation length. Acta Agric Scand. A Anim Sci. 2000;50:217–224.

23. Koketsu YG, Dial D, King VL. Influence of various factors on farrowing rate on farms using early weaning. J Anim Sci. 1997;75:580–587.

24. Buchanan DS, Johnson RK. Reproductive performance for four breeds of swine: crossbred females and purebred and crossbred boars. J Anim Sci. 1984;59:948–956.

25. Xue JL, Dial GD, Marsh WE, Davies PR, Momont HW. Influence of lactation length on sow productivity. Livest Prod Sci. 1993;34:253–265.

*26. Pattison HD, Cook GL, Mackenzie S. A study of culling patterns in commercial pig breeding herds [abstract]. Anim Prod. 1980;31:462–463.

27. Xue JL, Dial GD, Marsh WE, Lucia TL. Association between lactation length and sow reproductive performance and longevity. JAVMA. 1997;10:935–938.

28. Mabry J, Culbertson M, Reeves D. Effects of lactation length on weaning-to-first-service interval, first-service farrowing rate, and subsequent litter size. Swine Health Prod. 1996;4:185–188.

* Non-referred references.