Impact of Energy Intake and Pregnancy Status on Rate and Efficiency of Gain and

Transcription

1 CULL SOW FEEDING AND MANAGEMENT STRATEGIES Impact of Energy Intake and Pregnancy Status on Rate and Efficiency of Gain and Backfat Changes of Sows Post-weaning G. C. Shurson*,1, G. W. Libal**,, J. Crenshaw***,, C. R. Hamilton**,, R. L. Fisher*,, D. D. Koehler*,, and M. H. Whitney* 1 *Department of Animal Science, University of Minnesota, St. Paul, **Department of Animal and Range Science, South Dakota State University, Brookings 00, and ***Department of Animal and Range Sciences, North Dakota State University, Fargo ABSTRACT: A collaborative study was conducted to evaluate factors related to determining optimal feeding and management programs for increasing net returns from marketing cull sows. A total of multiparous sows averaging 1 kg body weight were weaned, moved to individual gestation crates, and assigned to one of 1 To whom corrrespondence should be addressed. Current address: 1 Riverstone Ct., College Station, TX. Current address: Darling International, 1 O Connor Ridge Blvd., Suite 00, Irving, TX 0. Current address: Vigortone Ag Products, Inc., Council St. NE, P.O. Box 0, Cedar Rapids, IA 0. Current address: AgriNutrition Services, Inc., 0 East Third Ave., Shakopee, MN. 1

2 eight treatment combinations in a x x factorial arrangement for a d postweaning feeding experiment. Factors included limit (L) (1. kg/sow/d) or ad libitum (AL) access to feed wk 1 post-weaning, corn-soybean meal (C) or barley-sunflower meal (B) diet, and pregnant (P) or non-pregnant (N) status. All sows were provided ad libitum access to feed from wk to wk post-weaning. Gain (G) and feed intake (FI) data were collected weekly for each sow and used to calculate gain:feed (G/F). Ultrasonic backfat (BF) data were collected on d 0, d 1, and d post-weaning. Sows on the AL treatment had greater FI (P < 0.0) but similar G (P = 0.0) for the d post-weaning period compared to sows on the L treatment. Most of this response was due to lower sow body weight loss during wk 1 post-weaning (P < 0.01) when sows were provided AL (-. kg) compared with L (-. kg) access to feed. Sows fed the C diet had higher G (P < 0.01), had improved G/F (P < 0.01) and increased BF (P < 0.01) over the d feeding period than sows fed B. The C diet resulted in less sow body weight loss (P < 0.01) during wk 1 (-. kg) than the B diet (-. kg). Pregnant sows (P) had higher G, FI, G/F and BF (P < 0.01) than N sows over the d feeding period. Most of this advantage occurred during wk postweaning when FI and G of N sows was lower (P < 0.01) than P sows. An economic analysis indicated that, when cull sow prices are relatively high and feed prices moderate to low, maintaining and managing cull sows for an additional weeks post-weaning may be economically advantageous compared to 0 or weeks. P sows fed the C diet produced the greatest economic return. These results suggest that mating sows as they return to estrus post-weaning and providing ad libitum access

3 to a corn-soybean meal diet improves growth performance and feed efficiency, and may thereby provide increased returns when marketing cull sows. Key Words: Swine, Sows, Post-weaning, Feed Intake, Pregnancy, Energy Introduction The sale of cull sows represents a small, but significant economic contribution ( to %) to the swine industry. Unfortunately, little research has been conducted to evaluate feeding, management and marketing strategies to improve the economic value of this derivative of pork production. Previous studies have shown variable performance levels of cull sows fed corn-soybean meal diets for various lengths of time post-weaning. Shurson et al. (11) showed that sow body weight and condition and previous level of performance contribute to expected performance of sows during a post-weaning feeding period. Sows can be either sold immediately after weaning as wet sows, or fed for a period of time and marketed as dry sows. Economic returns from feeding cull sows prior to market are dependent on a number of different factors including feed cost, rate and efficiency of weight gain, facility costs and/or availability, market price, and discounts for wet sows. Evaluating feeding programs and diet composition that may improve feed efficiency or weight gain may offer significant opportunities for swine producers to increase economic returns by feeding cull sows prior to

4 marketing. This study was conducted to test several post-weaning feeding strategies that may be implemented prior to marketing of cull sows. The objectives of this study were to evaluate the impact of dietary energy density, feeding level, and pregnancy status post-weaning on the rate, efficiency, and composition of weight gain in sows. In addition, an economic analysis was conducted to determine which strategy or combination of strategies provides the greatest net economic return, and if a or week post-weaning period provides additional return compared to marketing wet sows. Materials and Methods 1 Animals and Measurements A total of multiparous sows from three different herds, Univ. of Minnesota (MN), North Dakota State Univ. (ND), and South Dakota State Univ. (SD), were used. Genetic composition of sows at MN and SD were Yorkshire x Landrace F1's, and at ND were Yorkshire x Hampshire x Duroc females. Sows were individually housed in confinement gestation facilities and fed in individual gestation stalls throughout the experimental feeding period. Animals were weighed immediately following weaning and assigned to one of the eight treatments based on weight, body condition, and parity. All sows were

5 weighed weekly until conclusion of the d study to determine growth performance. Individual sow feed consumption was also measured on a weekly basis. Average backfat measurements were taken on all sows using a backfat probe (Renco LeanMeater ) at weaning, d 1 and d post-weaning in an attempt to measure body composition change. Sows assigned to pregnant (P) treatments were mated as they returned to estrus ( - d post-weaning) to provide comparative differences in rate, efficiency, and composition of gain. Sows that were mated but returned to estrus wk post-weaning were removed from the trial. Those sows assigned to restricted (R) feeding levels were limit fed 1. kg/d for the first wk post-weaning, and then allowed ad libitum feeding for the remaining wks of the trial. 1 Experimental Diets and Treatments All sows were fed either a corn-soybean meal (C) or barley-sunflower meal (B) based diet (Table 1) to study the effects of dietary energy density on growth rate, efficiency, and composition of gain. Separate vitamin and trace mineral premixes were used by each participating station, but all premixes were included to ensure that vitamins and minerals were provided to exceed NRC (1) minimum recommendations. Corn-soybean meal diets were formulated to contain kcal ME/kg, 1.% crude protein, and 0.% lysine. Sows were assigned to one of eight treatments arranged in a x x factorial of pregnancy status (pregnant (P) or non-pregnant (N)), feeding level during the first week post-weaning (ad libitum

6 (AL) or limit fed (L)), and diet composition (corn-soybean meal (C) or barleysunflower meal (B)). Barley-sunflower meal diets were formulated to contain kcal ME/kg,.% crude protein, and.% lysine. Diets were estimated to contain. and. % crude fiber, respectively, and were equivalent in Ca and P content. Statistical Analysis 1 1 The experiment was a x x factorial arranged in a randomized complete block design. The data were statistically analyzed using the GLM procedure of SAS (1). Individual sow measurements were considered the experimental unit. Least square means were calculated and used to evaluate responses to first week feeding level, diet fed (energy level), and pregnancy status for ADG, ADFI, G/F (weekly and total) and change in backfat (wk 1-, wk -, and wk 1-). Initial sow body weight at weaning was used as a covariate for ADG, ADFI, and G/F responses, while initial sow backfat at weaning was used as a covariate for evaluating changes in backfat depth. Location (station) did not appreciably affect (P > 0.) growth or backfat responses to first week feeding level, diet, or pregnancy status, and therefore those interactions were not included in the model. 1 0 Economic Analysis 1 An economic analysis was conducted utilizing growth performance and feed

7 1 intake data obtained from this experiment, based on the model used by Plain and Shurson (1). Least squares means for ADG and ADFI were utilized to determine body weight gain and feed intake over - and -week post-weaning periods, dependent on pregnancy status, feeding level week 1 post-weaning, diet fed, and all interactions between these dietary and management factors. An initial sow weaning weight of 10 kg and fixed non-feed cost of $1.0/week/sow marketed were assumed. A low or high ($ 0. and $0./kg) feed cost, low or high ($0.0 and $0.0/kg live body weight) cull sow market price, and presence or absence of a $0.0/kg wet sow market price discount were included in the model. Differences in diet cost, expense for breeding, and any additional non-feed costs were not considered, since these costs may be quite variable dependent on individual operation. A pregnant sow discount at the time of marketing was not considered, since most procurement facilities would not consider a difference at weeks gestation. In addition, many cull sows are marketed on a carcass yield basis, and therefore presence of conceptus would be accounted for. Results and Discussion 1 1 Sow Responses to Dietary Energy Density Post-Weaning 0 1 Sows fed the lower energy dense barley-sunflower meal diet (B) gained an average of 0.1 kg/sow/d less (P < 0.01) than sows fed the higher energy dense corn-

8 1 1 soybean meal diet (C) throughout the entire d post-weaning feeding period (Table ). Except during wks,, and, sows fed the C diet had increased daily gains throughout the trial, but especially during wk 1 (P < 0.01) and wk (P < 0.0). This resulted in an 1.% increase in body weight gain and.% increase in ending sow wt (P < 0.0). Feed intake was depressed for sows fed the lower energy diets during wk 1 and wk post-weaning (P < 0.01). Even though sows fed the B diet had a numerical decrease in feed intake (. kg/d) and energy consumption (.1 Mcal ME/d) than sows fed the C diet (. kg/d and 1. Mcal ME/d, respectively) over the d feeding period (Table ), they were also less efficient (0. kg gain/kg feed) compared to C sows (0.0 kg gain/kg feed) (P < 0.01) for the d study. This was due in part to lower feed efficiency rates during wks 1 (-0. vs. -0.1, P < 0.) and (0.1 vs. 0., P < 0.) post-weaning. Negative growth and feed efficiency values were observed for most sows during the first week post-weaning due to high levels of water loss during the post-suckling period. As expected, sows fed the lower energy B diet tended to accumulate less backfat from weaning to d 1 post-weaning (0. mm, P < 0.0), and over the entire d feeding period (1. mm, P < 0.01) compared to sows fed the higher energy C diet. 1

9 1 These results are in agreement with those reported from a previous study (Fernandez et al., 1) that showed that sows are relatively efficient in utilizing high fiber feedstuffs (barley and sunflower meal), compared to growing pigs. However, another study (Pollman et al., 1) indicated that as the fiber content of the diet increases, digestibility of fiber, energy, and nitrogen decreases when comparing diets containing sun-cured alfalfa, tall wheat straw, and corn-soybean meal. These two studies suggest that both source and level of dietary fiber need to be considered when developing feeding programs to optimize nutrient (energy) utilization in sows. Results from this study suggest that a lower energy, higher fiber diet, such as the barley-sunflower diet provided in this experiment, allows reasonable gains and efficiency of gain with less backfat gain over a d feeding period. However, the feeding of a corn-soybean meal diet results in heavier sows at the end of the feeding period. Sow Responses to Feeding Level the First Week Post-Weaning There were no differences in overall average daily gain between sows that were limit-fed (L) or full-fed (AL) the C or B diet during wk 1. During the wk 1 postweaning, limit-fed sows lost an average of 1. kg/d, while full-fed sows lost an average of 1.0 kg/d (Table ), a % reduction in body weight loss (P < 0.01). However, the.0 kg difference in total body weight loss between limit-fed and full-

10 1 fed sows at the end of week 1 was almost completely recovered at the end of the d experimental period (. vs.0 kg) (P > 0.0). This suggests that sows that are nutrient restricted during the first week post-weaning achieve compensatory gain during subsequent feeding periods. Sows that were limit-fed during the first week after weaning had similar increases in backfat accretion (. mm) during the d feeding period compared to sows that were full-fed during week 1 (. mm) (P = 0.). Sows that were limit-fed during the first week after weaning, however, ate. kg less feed (. vs. 0. kg) (P < 0.0) than full-fed sows during the entire d feeding period, and appeared to be slightly more efficient in converting feed to body weight gain (0.1 vs kg of gain/kg feed, respectively; P = 0.1). These results suggest there may be a slight advantage of limit feeding sows during the first week after weaning to reduce overall feed cost/kg of gain. Sow Responses to Pregnancy Status Post-Weaning Sows that were mated as they returned to estrus after weaning, and remained pregnant over the d feeding period (P), gained an average of 0. kg/sow/d more (P < 0.01) than sows that were not pregnant (N) during the study (Table ). Much of this weight gain advantage occurred during wk of the feeding period, when P sows gained 1.0 kg/d while N sows gained only 0. kg/d (P < 0.01). The minimal weight gain experienced by non-pregnant sows during week can be most likely

11 attributed to the fact that those sows returned to estrus during this week and consumed.0 kg less feed than pregnant sows (P < 0.01). However, the majority of the weight gain advantage of pregnant sows was due to higher average daily feed consumption (.0 vs.. kg/d) (P < 0.01) over the d feeding period. Pregnant sows also were more efficient (P < 0.01) in converting feed to body weight gain (0.0 kg gain/kg feed) than non-pregnant sows (0.1 kg gain/kg feed). However, pregnant sows had an increased (0%) rate of backfat deposition, resulting in a greater proportion of body weight gain being attributed to backfat deposition (P < 0.01). Nutrient utilization, body weight and compositional changes are different between pregnant and non-pregnant sows (Close et al., 1; Noblet et al., 1; Close et al., 1). In the pig there is a conflict of evidence on the extent to which maternal nutrition influences fetal growth. If there is a range of feed intakes over which fetal growth and development is independent of maternal nutrition, then there is likely to be a range in which nutrients are partitioned between maternal and reproductive tissues (Close et al., 1). Net maternal gain and the differences between pregnant and non-pregnant animals have been shown to vary with both feeding level and stage of gestation (Close et al., 1). These results may suggest that hormonal patterns associated with pregnancy have a large effect on body weight gain, ad libitum feed intake and gain efficiency, and that increased nutrient partitioning toward maternal growth appears to occur.

12 Sow Responses to Feeding Level, Dietary Energy Density, and Pregnancy Status Post- Weaning 1 1 Despite the fact that sows lose body weight after weaning while consuming considerable amounts of feed (Shurson et al., 11), feeding a low metabolizable energy density barley-sunflower meal diet, further increases body weight loss (Tables,, and ), regardless of whether sows are limit- or full-fed during the first week post-weaning (P < 0.01). As expected, no effect on weight loss during wk 1 post-weaning can be attributed to whether or not sows are mated as they return to estrus within that time period (Table ). Pregnancy status, however, significantly affects growth, feed intake and efficiency, and backfat accretion the first wks postweaning (Table ), regardless of diet (Table ) or feeding level wk 1 post-weaning (Table ) (P < 0.). One study has shown that loss of body water at weaning is considerable, but is significantly less than that which occurs in the fasted animal, and depletion of body tissue (including protein) occurs in the newly weaned sow (Zoiopoulos, 1). Therefore, in order to minimize sow weight loss during the first week post-weaning, sows should be fed a corn-soybean meal diet ad libitum. Mating sows further improves growth performance and feed conversion Economic Analysis of Different Manangement Regimes on Marketing Cull Sows at or Weeks Post-weaning 1

13 Results of an economic analysis of the results achieved from this trial are presented in Tables -. Net profitability per sow marketed is presented for each factor included, in addition to the estimated feed consumed and body weight gained that is assumed in the model. Including the presence of a wet sow discount of $0.0/kg live weight in the model resulted in an additional $1.1 to $. / cull sow marketed, dependent on feeding and/or management factors and feed costs and sow prices included, but did not guarantee a net profit for marketing cull sows at - or - weeks post-weaning. In general, feeding for weeks post-weaning was advantageous compared to weeks only when sow price was high ($0.0/kg) and feed cost low ($0.). When net return was near or below break-even, feeding for weeks was more economically advantageous than feeding for weeks. This indicates that if market prices are such that it is economically justified to feed cull sows, then a -week period is preferable to a -week period. Sows that were mated and pregnant, and sows that were fed the corn-soybean meal diet had the greatest economic returns. There was no appreciable difference in feeding level the first week post-weaning on net profitability. Combining the P and C factors resulted in additional increases in net profitability. However, this economic model doesn t consider a difference in feed cost between the C and B diets, nor does is consider the cost associated with breeding a sow. Although these costs should be relatively low, they will decrease the expected net return for each of these strategies. The least profitable scenarios occurred when sows were not mated (NP) and/or were fed the B diets.

14 Implications Ad libitum feeding during the first week post-weaning increased overall ADFI and tended to improve ADG and sow weight loss wk 1 post-weaning, but did not affect overall growth performance, sow body weight, or backfat accretion. Additionally, it did not appear to appreciably affect net return per cull sow marketed. Feeding a barley-sunflower meal diet decreased ADG and feed efficiency, and appeared to lower ADFI slightly. Compared to non-mated sows, sows that were mated immediately post-weaning had increased ADG, ADFI, G/F, and backfat levels for the d trial. Sows that were pregnant and/or fed a corn-soybean meal had the greatest net returns when fed for - or -weeks post-weaning. These results suggest that maximum growth and feed efficiency to increase the marketing weight of cull sows can be achieved by mating sows as they return to estrus post-weaning and providing a corn-soybean meal diet. Maximum net economic returns may be achieved by implementing this strategy for weeks post-weaning. Further improvements to feed efficiency may be achieved by restricting feed intake the first wk post-weaning. However, it is a combination of other factors, including feed cost and sow price (wet and dry), that determines whether to market cull sows wet, or which diet to be fed and for what period of time if marketing cull sows dry. 1

15 Literature Cited Close, W. H., J. Noblet, and R. P. Heavens. 1. The partition of body-weight gain in the pregnant sow. Livestock Prod. Sci. :-. Close, W. H., J. Noblet, and R. P. Heavens. 1. Studies on energy metabolism of the pregnant sow: The partition and utilization of metabolizable energy intake in pregnant and non-pregnant animals. Brit. J. Nutr. :-. Fernandez, J. A., H. Jorgensen, and A. Just. 1. Comparative digestibility experiments with growing pigs and adult sows. Anim. Prod. :-. Noblet, J., W. H. Close, and R. P. Heavens. 1. Studies on the energy metabolism of the pregnant sow: Uterus and mammary tissue development. Brit. J. Nutr. :1-. NRC. 1. Nutrient requirements of swine. th ed. National Academy Press, Washington, DC. Plain, R. L. and G. C. Shurson. 1. Marketing cull sows. Pork Ind. Hand. No.. Purdue Univ. Coop. Ext. Ser., West Lafayette, Ind. Pollman, D. S., D. M. Danielson, and E. R. Peo, Jr. 1. Value of high fiber diets for gravid swine. J. Anim. Sci. :-. SAS. 1. SAS User s Guide: Statistics. SAS Inst., Inc., Cary, NC. Shurson, G. C., G. A. Isler, K. M. Irvin, and G. A. Peterson. 11. Traits affecting post-weaning weight gain and feed intake of primiparous sows. J. Anim. Sci. :-.

16 Zoiopoulos, P. E., J. H. Topps, and P. R. English. 1. Losses in weight and body water in sows after weaning. Brit. J. Nutr. 0:-.

17 Table 1. Percentage composition of experimental diets (as-fed basis) Barley- Corn- Sunflower Soybean Item meal (B) a meal (C) b Ingredient, % Corn.0 Barley.0 Soybean meal (% CP).00 Sunflower meal (%CP) 1.0 Dicalcium phosphate Limestone Salt TMVP c a Formulated to.% CP, 0.% lysine, kcal ME/kg,.% crude fiber, 0.% Ca, and 0.0% P. b Formulated to 1.% CP, 0.% lysine, kcal ME/kg,.% crude 1 fiber, 0.% Ca, and 0.0% P. c Individual station trace mineral and vitamin premixes used, with varying levels substituted with grain. Biotin premix, choline chloride, and folic acid included at 0,, and g/ton, respectively.

18 Table. Least squares means comparison of pregnancy status, feeding level, and diet Non- Ad Limit Barley Corn Preg Preg Lib Fed Sun Soy Item (NP) (P) P -value (AL) (L) P -value (B) (C) P -value SEM No. sows Parity Sow wt, kg Farrow Wean (d 0) End (d ) ADG, kg Wk Wk Wk Wk Wk Wk Wk ADFI, kg Wk Wk Wk Wk Wk Wk Wk G/F Wk Wk Wk Wk Wk Wk Wk Chg in backfat, mm d d d

19 Table. Effects of a diet type x feeding level interaction on ADG, ADFI, G/F, and change in backfat thickness values of cull sows d 0 - post-weaning Corn-soybean meal (C) Barley-sunflower meal (B) Significance (P -value) Ad-libitum Limit Ad-libitum Limit CAL vs BAL vs CAL vs CL vs Item feeding (AL) feeding (L) feeding (AL) feeding (L) CL BL BAL BL ADG, kg ADFI, kg G/F BF change, mm

20 Table. Effects of a diet type x pregnancy status interaction on ADG, ADFI, G/F, and change in backfat thickness values of cull sows d 0 - post-weaning Corn-soybean meal (C) Barley-sunflower meal (B) Significance (P -value) Pregnant Non- Pregnant Non- CP vs BP vs CP vs CN vs Item (P) pregnant (N) (P) pregnant (N) CN BN BP BN ADG, kg ADFI, kg G/F BF change, mm

21 Table. Effects of a feeding level x pregnancy status interaction on ADG, ADFI, G/F, and change in backfat thickness values of cull sows d 0 - post-weaning Ad-libitum feeding (AL) Limited feeding (L) Significance (P -value) Pregnant Non- Pregnant Non- ALP vs LP vs ALP vs ALN vs Item (P) pregnant (N) (P) pregnant (N) ALN LN LP LN ADG, kg ADFI, kg G/F BF change, mm

22 Table. Economic analysis of different feeding and management regimens over a week post-weaning period, assuming no wet sow discount, on net profitability per cull sow marketed 1 Feeding Profit ($/sow) Feed Ending Sow Price = $0.0/kg Sow Price = $0.0/kg Trt Consumed Sow Wt Feed cost / kg Feed cost / kg Group (kg) (kg) $0. $0. $0. $0. Pregnant (P) Non-pregnant (NP) Ad libitum (AL) Limit-fed (L) Corn-soy (C) Barley-sun (B) P + AL P + L P + C P + B NP + AL NP + L NP + C NP + B AL + C AL + B L + C L + B P + AL + C P + AL + B P + L + C P + L + B NP + AL + C NP + AL + B NP + L + C NP + L + B Assumes initial weaned sow weight of 10 kg and nonfeed cost of $1.0/week/sow marketed.

23 Table. Economic analysis of different feeding and management regimens over a week post-weaning period, assuming a $ 0.0/kg live weight wet sow discount, on net profitability per cull sow marketed 1 Feeding Profit ($/sow) Feed Ending Sow Price = $0.0/kg Sow Price = $0.0/kg Trt Consumed Sow Wt Feed cost / kg Feed cost / kg Group (kg) (kg) $0. $0. $0. $0. Pregnant (P) Non-pregnant (NP) Ad libitum (AL) Limit-fed (L) Corn-soy (C) Barley-sun (B) P + AL P + L P + C P + B NP + AL NP + L NP + C NP + B AL + C AL + B L + C L + B P + AL + C P + AL + B P + L + C P + L + B NP + AL + C NP + AL + B NP + L + C NP + L + B Assumes initial weaned sow weight of 10 kg and nonfeed cost of $1.0/week/sow marketed.

24 Table. Economic analysis of different feeding and management regimens over a week post-weaning period, assuming no wet sow discount, on net profitability per cull sow marketed 1 Feeding Profit ($/sow) Feed Ending Sow Price = $0.0/kg Sow Price = $0.0/kg Trt Consumed Sow Wt Feed cost / kg Feed cost / kg Group (kg) (kg) $0. $0. $0. $0. Pregnant (P) Non-pregnant (NP) Ad libitum (AL) Limit-fed (L) Corn-soy (C) Barley-sun (B) P + AL P + L P + C P + B NP + AL NP + L NP + C NP + B AL + C AL + B L + C L + B P + AL + C P + AL + B P + L + C P + L + B NP + AL + C NP + AL + B NP + L + C NP + L + B Assumes initial weaned sow weight of 10 kg and nonfeed cost of $1.0/week/sow marketed.

25 Table. Economic analysis of different feeding and management regimens over a week post-weaning period, assuming a $ 0.0/kg live weight wet sow discount, on net profitability per cull sow marketed 1 Feeding Profit ($/sow) Feed Ending Sow Price = $0.0/kg Sow Price = $0.0/kg Trt Consumed Sow Wt Feed cost / kg Feed cost / kg Group (kg) (kg) $0. $0. $0. $0. Pregnant (P) Non-pregnant (NP) Ad libitum (AL) Limit-fed (L) Corn-soy (C) Barley-sun (B) P + AL P + L P + C P + B NP + AL NP + L NP + C NP + B AL + C AL + B L + C L + B P + AL + C P + AL + B P + L + C P + L + B NP + AL + C NP + AL + B NP + L + C NP + L + B Assumes initial weaned sow weight of 10 kg and nonfeed cost of $1.0/week/sow marketed.