Lactational Performance of High Producing Dairy Cows Fed Diets Containing Salmon Meal and Urea' PAUL M. WINDSCHITL Agricultural and Forestry Experiment Station University of Alaska Palmer 99E45
decrease in milk fat percentage and yield without any beneficial effects on milk production or lactational efficiency. (Key words: fish meal, undegraded intake protein, lactational response)
ABSTRACT
Thirty Holstein cows were used in a 12-wk trial to study the effects of salmon meal and urea on lactational performance. Two experimental diets, one containing 5.6% salmon meal and the other 5.2% salmon meal plus .42% urea, were compared with a soybean meal control diet. Salmon meal and urea replaced a portion of the soybean meal. Dietary undegraded intake protein levels (expressed as percentage of CP) were 28.8, 35.6, and 32.4% for soybean meal, salmon meal, and salmon meal plus urea Total mixed diets (average 17.3% CP, 17.6% ADF) consisting of 60% concentrate mixture and 40% bromegrass silage (DM basis) were fed twice daily. Total DMI was lower with salmon meal compared with soybean meal (20.2 versus 22.2 kg/ d); salmon meal plus urea (21.2 kg/d) was intermediate. Actual milk production was similar for all diets (average 41.1 kg/d). Percentage milk fat and 4% FCM yield were lower with salmon meal (2.56%,31.6 kg/d) and salmon meal plus urea (2.50%, 31.4 kg/d) than with soybean meal (3.03%,35.9 kg/d). Gross efficiency (weight FCM/weight DMI) was higher for soybean meal than for salmon meal and salmon meal plus urea Acetate:propionate tended to be higher with the soybean meal diet. The use of a high oil fish meal to provide a source of rumen undegraded intake protein, alone or in combination with urea, resulted in a
Abbreviation key: FAME = fatty acid methyl esters, PUN = plasma urea N, SBM = soybean meal, SM = salmon meal, UIP = undegraded intake protein, USM = urea plus salmon meal. INTRODUCTION
The use of low rumen-degradable proteins in diets of high producing dairy cows has received considerable attention in anticipation of enhanced lactational performance. Attempts to alter the amount of undegraded intake protein (UIP)in the diet have included the use of chemically or physically treated protein sources (17, 26) and the use of naturally low degradable proteins such as fish meal and other animal by-product meals (7, 17). Fish meal is an attractive protein s o m e in that it is low in degradabjJity and also is relatively well balanced in AA content. Compared with soybean meal (SBM), fish meal is higher in lysine and methionine, two AA often implicated as limiting optimal milk production (17). Fish meal also contains a higher proportion of UIP than plant protein sources such as SBM (13). Increased duodenal flow of essential AA, particularly lysine and methionine, has been reported in sheep when fish meal was included in the diet (11). Lactational performance was improved when fish and blood meals were used to increase the proportion of UIP in the total diet of dairy cows (9). Several other researchers (2, 19) reported no response in milk yield and a Received November 13, 1990. Accepted May 31, 1991. decrease in milk fat percentage when fish meal hblished with the approval of the director as Paper was used to provide UIP for early lactation Number J-216,Journal Series, Alaska Agricultaral and Porcsby Experiment Station, University of Alaska, Pair- cows. The decrease in milk fat percentage and yield may be due to the residual fish oil found banks. 1991 J Dairy Sci 74:3475-3485
3475
3476
WINDSCIillI
in the meal. Fish oil contains significant quantities of longchain polyunsaturated fatty acids, which have been shown to impact negatively the uptake of plasma fatty acids by the mammary gland (22). In addition, Hoover et aL (IO) presented evidence to suggest that fish oil had a direct effect on m e n fermentation and lowered acetate:propionate, possibly decreasing the available milk fat precursors. Polyunsaturated fatty acids may be toxic to certain cellulolytic bacteria of the rumen, resulting in decreassd fiber digestibility (16). One strategy for using low degradable proteins may be to use them in combination with readily available NPN sources such as urea (17). The N H 3 derived from uma could meet the needs of the rumen microorganisms. Beneficial effects have been reported (7)when NPN was added to diets containing low degradable proteins. In particular, Erasmus et al. (7) reported that urea could be used successfully in lactating cow diets containing fish meal. The objective of this experiment was to evaluate the effects of substituting fish meal with or without urea for SBM in high producing dairy cow diets on lactational performance and milk composition. MATERIALS AND METHODS
Animals and Treatment
The fish meal used in this study was derived from the processing wastes of salmon fisheries (Icicle Seafoods, Sewad, AK); It contained, on average, 61.4% CP, 13.8% ether extract, 17% ash, 5.5% Ca, and 3.2% P. The U P content (expressed as percentage of total CP) of the salmon meal (SM) as determined by the in situ technique was 62%. Compared with NRC (13) values for menhaden meal, the SM was higher in ether extract (13.8 versus 10.5%) and lower in CP (61.4 versus 66.7%). Fatty acid content of the SM oil is given in Table 1. Thirty early lactation Holstein cows were used in a 12-wk experiment involving the feeding of SBM, SM,or a combination of SM plus urea (USM). The three treatments were SBM, SM, and USM. "he average lactation number for cows in each of the three treatment groups was 3.1, 3.0, and 3.0 for SBM, SM, and USM, respectively. All cows were 22 DIM Journal of Dairy Science Vol. 74. No. 10, 1991
TABLE 1. Fatly acid content of salmon meal oil.
Faav acid'
woo Id 14:O
6.69 .36 22.62 6.88 2.03 4.69 25.23 2.14 1.39 10.46 103 4.01 3.22
15:o
16:O 161 n-7 17:O l8:O 18:l n-9 18:2 n-6 18:3 n-3 2&1 n-9 22:l -11 205 n-3 226 n-3 ~
~
~
~~
~
'Denoted as numbex of carb0ns:rmmbcr of double bonds; XI-x, whm x is the first unsaturated carbon from the methyl end.
at the beginning of experimental periods. Previws lactation 305d milk production averages for cows in each of the treatment groups were 7710,7358, and 7579 kg for SBM, SM, and USM. Because individual cows were started on trial over a 6-mo period, cows were blocked by calving date and assigned randomly within calving date p u p to one of the three treatment groups. Pretrial milk production averages were 38.5, 39.7,and 39.6 kg/d for SBM, SM, and USM. Cows were housed in individual pens starting wk 2 postpartum. Weeks 2 to 3 postpartum were used as a covariate period for collection of pretreatment data on milk production and composition. The experimental period was wk 4 through 15. Ingredient composition of the pelleted concentrate mixtures is given in Table 2. The SM and USM diets provided approximately 1.3 and 1.1 kg/d of SM, respectively. These levels of SM were selected based on prdiminary research at the University of Alaska (L. B. Bruce, unpublished data). All cows were fed a TMR twice M y consisting of 60% concentrate mix and 40% bromegrass silage (DM basis). Concentrate pellets and silage were mixed together in a stationary horizontal batch mixer (H. C. Davis Sons Co., Inc., Bonner Springs, KS). Rations were mixed separately for individual cows. Dry matter content of the silage was determined weekly by oven drying at 55'C, and diets were adjusted as necessary
SALMON MEAL AND UREA
3477
TABLE 2. Ingredient content of pelleted concentrate mix-
were determined according to AOAC (1) pro-
tares.1
cedures.Neutral detergent fiber and ADF were
determined sequentially by the method of Van Soest and Robertson (24). Protein degradabilhgdient SBM SM USM ity of the concentrate mixtures was determined (% of DM) by the in situ technique as described by Wind15.6 15.6 15.6 Barley schitl and Stem (26). The rumen-fistulated cow 43.3 51.3 54.7 COm used for the in situ work was fed the SBM Alfalfa meal 9.4 9.4 9.4 diet. Animal fat 3.9 2.0 2.1 1.6 1.6 1.6 Molasses Cows were milked twice daily with milk Soybean meal 22.9 9.1 5.5 weights recorded at each milking. During wk Salmon meal ... 9.4 8.6 3, two 24-h milk samples were collected (a.m. ... ... .7 Urea and p.m.) from each cow for use as covariate Limestone 1.5 .8 .9 data. Milk was sampled weekly throughout the Dicalcium phosphate 1.0 ... .1 Trace-mineralizcd salt3 .8 .8 .8 experimental period. Samples were analyzed ~ S B M= soybtan meat; SM = salmon wal; USM = for fat, protein, lactose, and SNF by infr;ued spectrophotometry (Multispec M IR spectrosalmon meal plus urea. kontains 6600 IUvitamin f i g , 2200 TU vitamin D/ photometer; Foss Food Technology Corp., kg. 33 Iu vitamin E/Lg. Eden Prairie, MN). Milk samples collected 3Composition: 98% .NaCl, .35% Zn, .28% Mn, .175% during wk 8 and 12 were freezedrid and Fe. .035% Cu, .007% I, .OW% Co. stored for analysis of fatty acid distribution. Milk fat was extracted from the freeze-dried samples by the Mojonnier procedure (1). Fatty acid methyl esters (FAME) were prepared acto maintain the forage to concentrate ratio. cording to the procedure of Metcalfe et al. During the covariate period, all cows received (12). Analysis of FAME was performed on a the SBM diet. Cows receiving the SM and Perkin-Elmer model 8320 capillary gas c h r e USM diets were switched over a 5 4 period matograph (Perkin-Elmer Corporation, Norfrom the SBM diet to their respective experi- walk, Cl') using an SP-2330 fused silica (high mental diets beginning the latter part of wk 3, polarity cyanosilicone phase), 30-m x .32-mm after the preliminary milk data were collected. i.d. capillary column (Supelco, Inc., All cows were receiving their full experimental Bellefonte, PA). Operating conditions were as diets beginning wk 4. Body weights were re- follows: column temperature, 8 min at 150°C. corded on 3 consecutive d at the beginning and then to 1 W C at 3'C/min and held for 15 min; end of the experimental period and once every injector and detector temperatures, 200 and 250'C, respectively; linear velocity, 20 cm/s, 2 wk during the trial. helium; flame ionization detectoq sample size, 1 pl; split ratio, 1OO:l. Individual fatty acids Sample Collection and Analysis were identified based on column retention Feed refusals were recorded daily to deter- times of hown fatty acid standards of both mine daily DMI. Concentrate mixture samples marine and animal sources. were collected weekly and composited by Ruminal contents and jugular vein blood month for chemical analyses; silage samples samples were obtained during wk 8 and 12, were collected weekly and analyzed by week. approximately 2 h after the am. feeding. RuFeed samples were oven dried at 55'C for 48 h men fluid was collected by applying vacuum and ground through a 1-mm Wdey mill (Ar- to an esophageal tube fitted with a stainless thur H. Thomas. Philadelphia, PA) screen for steel suction strainer (Precision Machine Co., subsequent analyses. A 1-g sample of the Inc., Lincoln, NE). Measurements of pH were ground feeds was dried at 1M'C for 24 h to taken immediately after collection of the rudetermine absolute DM percentage. Organic men fluid. Rumen fluid (50 ml) then was matter was determined after samples were placed into plastic vials containing 2 ml of ashed in a muffle furnace at 550'C for 12 h. 50% sulfuric acid to retard microbial activity Crude protein, ether extract, Ca, P, and Mg further. Acidified samples were frozen until concentrate
-
mix2
-
Journal of Dairy Science Vol. 74, No. 10, 1991
3478
WINDSCHlTL
TABLE 3. Chemical composition of concentratemixtures, forages, and TMR for cows fed diets containing soybean meal (SBM). salmon meal (SM), or salmon meal plus urea (USM). ~~~~~
~
~~
~
coocenbrate mix
Measurement
SBM
SM
USM
DM, 96
87.5
87.7
87.4
OM CP
93.6 20.0 28.7 135 6.9 6.6 1.14 58 21
93.6 19.8 38.6 13.6 5.9 6.2 1.16
93.4 20.1 33.9 13.9 6.5 6.6 1.32
.18
.18
UIP? 46 of CP NDP
ADF Ether extract Ca P
Ms
.a
.66
Bromegrass silage 35.4 (96 of DM) 92.6 13.3 29.0 60.0 34.2 4.1 .47 26 .20
TOM
diet'
SBM
SM
USM
...
...
...
932 17.3 28.8 32.1 17.8 5.6 .87 .45 21
93.2 172 35.6 32 2 17.2 5.4 .a8
93.1 17.4 32.4 32.3 17.6 5.6 .98
SO
50
.19
.19
1mdated. 2Undegraded intake proteia The UIP values for conrentrates were determiged by in situ technique. The UIP value for bromepass silage is taken from NRC (13) estimate for grass silage.
analyzed for N H 3 by steam distillation (3) and VFA content by gas chromatography. Rumen fluid samples were prepared for VPA analysis using the method of Erwin et al. (8). Volatile fatty acids were analyzed on a Perkin-Elmer Model 8320 capillary gas chromatograph using a Nukol fused silica 30-m x .25-mm i.d. capillary column (Supelco, Inc., Bellefonte, PA). Operating conditions were as follows: column temperature, 182'C; injector and detector temperatures, 232 and 292'C, respectively; hear velocity, 20 c d s , helium; flame ionization detector; sample size, 1 pl; split ratio 1OO:l. Samples of jugular vein blood were drawn into heparinized vacutainer tubes at the time of rumen sampling. Samples were centrihged at 10oO x g for 15 min at 1o'C to separate plasma. The separated plasma was removed from the vacutainer tube using a serum filter isolator (Iso-filter, Becton-Dickinson Co.,Oxnard, CA). Plasma was from until analyzed for urea N (Sigma urea N kit Number 640, Sigma Chemical Co., St. Louis, MO). Statlstlcal Analysis
Data were analyzed as a completely randomized block design. Milk production and composition data were adjusted by covariance analysis (21) using milk data from wk 3 as the covariates. Data were analyzed by least squares using the general linear models proceJournal of Dairy Scicncc Vol. 74. No. 10, 1991
dure of SAS (SAS Institute, Inc., Cary, NC). Results are expressed as least squares means. Fisher's least significant difference test (21) was used to separate treatment least squares means when a significant F test (P < .05) occurred. Differences between treatment means were declared significant at P c .05 unless noted otherwise. RESULTS AND DISCUSSION
Chemical Composltlon of Feeds
Chemical composition of the concentrate mixtures and bromegrass silage is listed in
Table 3. The TMR were balanced to be isonitrogenous and isocaloric. Based on nutrient content of the individual concentrate and forage components, actual nutrient content of the TMR was calculated. Concentrate mixtures were similar in CP content but varied in the amount of degradable protein (Table 4). The ADF content of the diets was lower than recommended by NRC (13) due to the lower than expected ADF content of the bromegrass silage. Ether extract content of the diets was similar. Animal fat (tallow) was added to increase the energy density of the diets and also to balance for the presence of oil in the SM to keep diets isocaloric. The average amount of fish oil supplied by the SM and USM diets was 180 and 152 g/d, respectively. Magnesium
3479
SALMON MEAL AND UREA TABLE 4. In situ estimation of concentmte mixture protein and DM degradabdity and rate of digestion.' concentrate mix
TABLE 5. Dry matter intake and BW for cows fed diets Containing soybean meal (SBM), salmon meal, (SM), or salmon meal plus urea (USM).
Diet
~
SBM
Item
Protein degradation,* % 71.3 N Remaining at 24 h, 96 11.6 N Rate of digestionb .074 DM Remaining at 24 h, % 7.9 DM Rate of digestionb .095
SM 61.4 23.6 .043 15.7
.064
USM 66.1
23.0 .042 15.6
.066
%BM = soybean m d , SM = &on mea^; USM = salmon meal plus urea. 2Determined at Kr = .OS& where Kris the rate conmant for passage of undegraded CP from the rumen.
content of the diets was slightly below the level recommended for early lactation cows (13). Dry Matter intake and Body Weights
Dry matter intake (weight per day and as a percentage of BW) tended to be lower with the diets containing SM (Table 5); the difference between the SM and SBM diets was significant. Dry matter intake throughout the trial (Figure 1) was consistently higher with the SBM diet. The apparent drop in DMI on the SM diet during wk 9 to wk 12 cannot be explained. Dry matter intake curves for the SM and USM diets tended to be more erratic than for the SBM diet. The addition of urea to the USM diet appeared to have a positive effect on DMI in comparison with the SM diet. Overall, there were no obvious problems with regard to feed refusals. Other researchers also reported decreases in DMI when fish meal was included in the diet of lactating animals (15). In another study, Spain et al. (19) reported no change in DMI when fish meal was fed to midlactation dairy cows. In general, it appears that DMI may be reduced slightly when fish meal is included in the diet. Although the fish meal odor may be objectionable to some animals, the use of a TMR may help to alleviate any serious problems with feed refusals. Although DMI was lower with the diets containing SM, BW changes (Table 5) were not significantly affected. A tendency was observed for slightly higher weight gains on the diets containing SM. This may be a result of less energy going toward the production of
:
:
SBM
Item
22.2' 3.6' Average BW, kg 620 SWChange,kg/wk 1.9 DMI, W d DMI, 46 of BW
SM 2O.Zb 3.3b 613 2.8
USM 21.2&
3.5&
608 3.0
+"Means in same row with unlike superscripts
(P
decrease in milk fat percentage and yield without any beneficial effects on milk production or lactational efficiency. (Key words: fish meal, undegraded intake protein, lactational response)
ABSTRACT
Thirty Holstein cows were used in a 12-wk trial to study the effects of salmon meal and urea on lactational performance. Two experimental diets, one containing 5.6% salmon meal and the other 5.2% salmon meal plus .42% urea, were compared with a soybean meal control diet. Salmon meal and urea replaced a portion of the soybean meal. Dietary undegraded intake protein levels (expressed as percentage of CP) were 28.8, 35.6, and 32.4% for soybean meal, salmon meal, and salmon meal plus urea Total mixed diets (average 17.3% CP, 17.6% ADF) consisting of 60% concentrate mixture and 40% bromegrass silage (DM basis) were fed twice daily. Total DMI was lower with salmon meal compared with soybean meal (20.2 versus 22.2 kg/ d); salmon meal plus urea (21.2 kg/d) was intermediate. Actual milk production was similar for all diets (average 41.1 kg/d). Percentage milk fat and 4% FCM yield were lower with salmon meal (2.56%,31.6 kg/d) and salmon meal plus urea (2.50%, 31.4 kg/d) than with soybean meal (3.03%,35.9 kg/d). Gross efficiency (weight FCM/weight DMI) was higher for soybean meal than for salmon meal and salmon meal plus urea Acetate:propionate tended to be higher with the soybean meal diet. The use of a high oil fish meal to provide a source of rumen undegraded intake protein, alone or in combination with urea, resulted in a
Abbreviation key: FAME = fatty acid methyl esters, PUN = plasma urea N, SBM = soybean meal, SM = salmon meal, UIP = undegraded intake protein, USM = urea plus salmon meal. INTRODUCTION
The use of low rumen-degradable proteins in diets of high producing dairy cows has received considerable attention in anticipation of enhanced lactational performance. Attempts to alter the amount of undegraded intake protein (UIP)in the diet have included the use of chemically or physically treated protein sources (17, 26) and the use of naturally low degradable proteins such as fish meal and other animal by-product meals (7, 17). Fish meal is an attractive protein s o m e in that it is low in degradabjJity and also is relatively well balanced in AA content. Compared with soybean meal (SBM), fish meal is higher in lysine and methionine, two AA often implicated as limiting optimal milk production (17). Fish meal also contains a higher proportion of UIP than plant protein sources such as SBM (13). Increased duodenal flow of essential AA, particularly lysine and methionine, has been reported in sheep when fish meal was included in the diet (11). Lactational performance was improved when fish and blood meals were used to increase the proportion of UIP in the total diet of dairy cows (9). Several other researchers (2, 19) reported no response in milk yield and a Received November 13, 1990. Accepted May 31, 1991. decrease in milk fat percentage when fish meal hblished with the approval of the director as Paper was used to provide UIP for early lactation Number J-216,Journal Series, Alaska Agricultaral and Porcsby Experiment Station, University of Alaska, Pair- cows. The decrease in milk fat percentage and yield may be due to the residual fish oil found banks. 1991 J Dairy Sci 74:3475-3485
3475
3476
WINDSCIillI
in the meal. Fish oil contains significant quantities of longchain polyunsaturated fatty acids, which have been shown to impact negatively the uptake of plasma fatty acids by the mammary gland (22). In addition, Hoover et aL (IO) presented evidence to suggest that fish oil had a direct effect on m e n fermentation and lowered acetate:propionate, possibly decreasing the available milk fat precursors. Polyunsaturated fatty acids may be toxic to certain cellulolytic bacteria of the rumen, resulting in decreassd fiber digestibility (16). One strategy for using low degradable proteins may be to use them in combination with readily available NPN sources such as urea (17). The N H 3 derived from uma could meet the needs of the rumen microorganisms. Beneficial effects have been reported (7)when NPN was added to diets containing low degradable proteins. In particular, Erasmus et al. (7) reported that urea could be used successfully in lactating cow diets containing fish meal. The objective of this experiment was to evaluate the effects of substituting fish meal with or without urea for SBM in high producing dairy cow diets on lactational performance and milk composition. MATERIALS AND METHODS
Animals and Treatment
The fish meal used in this study was derived from the processing wastes of salmon fisheries (Icicle Seafoods, Sewad, AK); It contained, on average, 61.4% CP, 13.8% ether extract, 17% ash, 5.5% Ca, and 3.2% P. The U P content (expressed as percentage of total CP) of the salmon meal (SM) as determined by the in situ technique was 62%. Compared with NRC (13) values for menhaden meal, the SM was higher in ether extract (13.8 versus 10.5%) and lower in CP (61.4 versus 66.7%). Fatty acid content of the SM oil is given in Table 1. Thirty early lactation Holstein cows were used in a 12-wk experiment involving the feeding of SBM, SM,or a combination of SM plus urea (USM). The three treatments were SBM, SM, and USM. "he average lactation number for cows in each of the three treatment groups was 3.1, 3.0, and 3.0 for SBM, SM, and USM, respectively. All cows were 22 DIM Journal of Dairy Science Vol. 74. No. 10, 1991
TABLE 1. Fatly acid content of salmon meal oil.
Faav acid'
woo Id 14:O
6.69 .36 22.62 6.88 2.03 4.69 25.23 2.14 1.39 10.46 103 4.01 3.22
15:o
16:O 161 n-7 17:O l8:O 18:l n-9 18:2 n-6 18:3 n-3 2&1 n-9 22:l -11 205 n-3 226 n-3 ~
~
~
~~
~
'Denoted as numbex of carb0ns:rmmbcr of double bonds; XI-x, whm x is the first unsaturated carbon from the methyl end.
at the beginning of experimental periods. Previws lactation 305d milk production averages for cows in each of the treatment groups were 7710,7358, and 7579 kg for SBM, SM, and USM. Because individual cows were started on trial over a 6-mo period, cows were blocked by calving date and assigned randomly within calving date p u p to one of the three treatment groups. Pretrial milk production averages were 38.5, 39.7,and 39.6 kg/d for SBM, SM, and USM. Cows were housed in individual pens starting wk 2 postpartum. Weeks 2 to 3 postpartum were used as a covariate period for collection of pretreatment data on milk production and composition. The experimental period was wk 4 through 15. Ingredient composition of the pelleted concentrate mixtures is given in Table 2. The SM and USM diets provided approximately 1.3 and 1.1 kg/d of SM, respectively. These levels of SM were selected based on prdiminary research at the University of Alaska (L. B. Bruce, unpublished data). All cows were fed a TMR twice M y consisting of 60% concentrate mix and 40% bromegrass silage (DM basis). Concentrate pellets and silage were mixed together in a stationary horizontal batch mixer (H. C. Davis Sons Co., Inc., Bonner Springs, KS). Rations were mixed separately for individual cows. Dry matter content of the silage was determined weekly by oven drying at 55'C, and diets were adjusted as necessary
SALMON MEAL AND UREA
3477
TABLE 2. Ingredient content of pelleted concentrate mix-
were determined according to AOAC (1) pro-
tares.1
cedures.Neutral detergent fiber and ADF were
determined sequentially by the method of Van Soest and Robertson (24). Protein degradabilhgdient SBM SM USM ity of the concentrate mixtures was determined (% of DM) by the in situ technique as described by Wind15.6 15.6 15.6 Barley schitl and Stem (26). The rumen-fistulated cow 43.3 51.3 54.7 COm used for the in situ work was fed the SBM Alfalfa meal 9.4 9.4 9.4 diet. Animal fat 3.9 2.0 2.1 1.6 1.6 1.6 Molasses Cows were milked twice daily with milk Soybean meal 22.9 9.1 5.5 weights recorded at each milking. During wk Salmon meal ... 9.4 8.6 3, two 24-h milk samples were collected (a.m. ... ... .7 Urea and p.m.) from each cow for use as covariate Limestone 1.5 .8 .9 data. Milk was sampled weekly throughout the Dicalcium phosphate 1.0 ... .1 Trace-mineralizcd salt3 .8 .8 .8 experimental period. Samples were analyzed ~ S B M= soybtan meat; SM = salmon wal; USM = for fat, protein, lactose, and SNF by infr;ued spectrophotometry (Multispec M IR spectrosalmon meal plus urea. kontains 6600 IUvitamin f i g , 2200 TU vitamin D/ photometer; Foss Food Technology Corp., kg. 33 Iu vitamin E/Lg. Eden Prairie, MN). Milk samples collected 3Composition: 98% .NaCl, .35% Zn, .28% Mn, .175% during wk 8 and 12 were freezedrid and Fe. .035% Cu, .007% I, .OW% Co. stored for analysis of fatty acid distribution. Milk fat was extracted from the freeze-dried samples by the Mojonnier procedure (1). Fatty acid methyl esters (FAME) were prepared acto maintain the forage to concentrate ratio. cording to the procedure of Metcalfe et al. During the covariate period, all cows received (12). Analysis of FAME was performed on a the SBM diet. Cows receiving the SM and Perkin-Elmer model 8320 capillary gas c h r e USM diets were switched over a 5 4 period matograph (Perkin-Elmer Corporation, Norfrom the SBM diet to their respective experi- walk, Cl') using an SP-2330 fused silica (high mental diets beginning the latter part of wk 3, polarity cyanosilicone phase), 30-m x .32-mm after the preliminary milk data were collected. i.d. capillary column (Supelco, Inc., All cows were receiving their full experimental Bellefonte, PA). Operating conditions were as diets beginning wk 4. Body weights were re- follows: column temperature, 8 min at 150°C. corded on 3 consecutive d at the beginning and then to 1 W C at 3'C/min and held for 15 min; end of the experimental period and once every injector and detector temperatures, 200 and 250'C, respectively; linear velocity, 20 cm/s, 2 wk during the trial. helium; flame ionization detectoq sample size, 1 pl; split ratio, 1OO:l. Individual fatty acids Sample Collection and Analysis were identified based on column retention Feed refusals were recorded daily to deter- times of hown fatty acid standards of both mine daily DMI. Concentrate mixture samples marine and animal sources. were collected weekly and composited by Ruminal contents and jugular vein blood month for chemical analyses; silage samples samples were obtained during wk 8 and 12, were collected weekly and analyzed by week. approximately 2 h after the am. feeding. RuFeed samples were oven dried at 55'C for 48 h men fluid was collected by applying vacuum and ground through a 1-mm Wdey mill (Ar- to an esophageal tube fitted with a stainless thur H. Thomas. Philadelphia, PA) screen for steel suction strainer (Precision Machine Co., subsequent analyses. A 1-g sample of the Inc., Lincoln, NE). Measurements of pH were ground feeds was dried at 1M'C for 24 h to taken immediately after collection of the rudetermine absolute DM percentage. Organic men fluid. Rumen fluid (50 ml) then was matter was determined after samples were placed into plastic vials containing 2 ml of ashed in a muffle furnace at 550'C for 12 h. 50% sulfuric acid to retard microbial activity Crude protein, ether extract, Ca, P, and Mg further. Acidified samples were frozen until concentrate
-
mix2
-
Journal of Dairy Science Vol. 74, No. 10, 1991
3478
WINDSCHlTL
TABLE 3. Chemical composition of concentratemixtures, forages, and TMR for cows fed diets containing soybean meal (SBM). salmon meal (SM), or salmon meal plus urea (USM). ~~~~~
~
~~
~
coocenbrate mix
Measurement
SBM
SM
USM
DM, 96
87.5
87.7
87.4
OM CP
93.6 20.0 28.7 135 6.9 6.6 1.14 58 21
93.6 19.8 38.6 13.6 5.9 6.2 1.16
93.4 20.1 33.9 13.9 6.5 6.6 1.32
.18
.18
UIP? 46 of CP NDP
ADF Ether extract Ca P
Ms
.a
.66
Bromegrass silage 35.4 (96 of DM) 92.6 13.3 29.0 60.0 34.2 4.1 .47 26 .20
TOM
diet'
SBM
SM
USM
...
...
...
932 17.3 28.8 32.1 17.8 5.6 .87 .45 21
93.2 172 35.6 32 2 17.2 5.4 .a8
93.1 17.4 32.4 32.3 17.6 5.6 .98
SO
50
.19
.19
1mdated. 2Undegraded intake proteia The UIP values for conrentrates were determiged by in situ technique. The UIP value for bromepass silage is taken from NRC (13) estimate for grass silage.
analyzed for N H 3 by steam distillation (3) and VFA content by gas chromatography. Rumen fluid samples were prepared for VPA analysis using the method of Erwin et al. (8). Volatile fatty acids were analyzed on a Perkin-Elmer Model 8320 capillary gas chromatograph using a Nukol fused silica 30-m x .25-mm i.d. capillary column (Supelco, Inc., Bellefonte, PA). Operating conditions were as follows: column temperature, 182'C; injector and detector temperatures, 232 and 292'C, respectively; hear velocity, 20 c d s , helium; flame ionization detector; sample size, 1 pl; split ratio 1OO:l. Samples of jugular vein blood were drawn into heparinized vacutainer tubes at the time of rumen sampling. Samples were centrihged at 10oO x g for 15 min at 1o'C to separate plasma. The separated plasma was removed from the vacutainer tube using a serum filter isolator (Iso-filter, Becton-Dickinson Co.,Oxnard, CA). Plasma was from until analyzed for urea N (Sigma urea N kit Number 640, Sigma Chemical Co., St. Louis, MO). Statlstlcal Analysis
Data were analyzed as a completely randomized block design. Milk production and composition data were adjusted by covariance analysis (21) using milk data from wk 3 as the covariates. Data were analyzed by least squares using the general linear models proceJournal of Dairy Scicncc Vol. 74. No. 10, 1991
dure of SAS (SAS Institute, Inc., Cary, NC). Results are expressed as least squares means. Fisher's least significant difference test (21) was used to separate treatment least squares means when a significant F test (P < .05) occurred. Differences between treatment means were declared significant at P c .05 unless noted otherwise. RESULTS AND DISCUSSION
Chemical Composltlon of Feeds
Chemical composition of the concentrate mixtures and bromegrass silage is listed in
Table 3. The TMR were balanced to be isonitrogenous and isocaloric. Based on nutrient content of the individual concentrate and forage components, actual nutrient content of the TMR was calculated. Concentrate mixtures were similar in CP content but varied in the amount of degradable protein (Table 4). The ADF content of the diets was lower than recommended by NRC (13) due to the lower than expected ADF content of the bromegrass silage. Ether extract content of the diets was similar. Animal fat (tallow) was added to increase the energy density of the diets and also to balance for the presence of oil in the SM to keep diets isocaloric. The average amount of fish oil supplied by the SM and USM diets was 180 and 152 g/d, respectively. Magnesium
3479
SALMON MEAL AND UREA TABLE 4. In situ estimation of concentmte mixture protein and DM degradabdity and rate of digestion.' concentrate mix
TABLE 5. Dry matter intake and BW for cows fed diets Containing soybean meal (SBM), salmon meal, (SM), or salmon meal plus urea (USM).
Diet
~
SBM
Item
Protein degradation,* % 71.3 N Remaining at 24 h, 96 11.6 N Rate of digestionb .074 DM Remaining at 24 h, % 7.9 DM Rate of digestionb .095
SM 61.4 23.6 .043 15.7
.064
USM 66.1
23.0 .042 15.6
.066
%BM = soybean m d , SM = &on mea^; USM = salmon meal plus urea. 2Determined at Kr = .OS& where Kris the rate conmant for passage of undegraded CP from the rumen.
content of the diets was slightly below the level recommended for early lactation cows (13). Dry Matter intake and Body Weights
Dry matter intake (weight per day and as a percentage of BW) tended to be lower with the diets containing SM (Table 5); the difference between the SM and SBM diets was significant. Dry matter intake throughout the trial (Figure 1) was consistently higher with the SBM diet. The apparent drop in DMI on the SM diet during wk 9 to wk 12 cannot be explained. Dry matter intake curves for the SM and USM diets tended to be more erratic than for the SBM diet. The addition of urea to the USM diet appeared to have a positive effect on DMI in comparison with the SM diet. Overall, there were no obvious problems with regard to feed refusals. Other researchers also reported decreases in DMI when fish meal was included in the diet of lactating animals (15). In another study, Spain et al. (19) reported no change in DMI when fish meal was fed to midlactation dairy cows. In general, it appears that DMI may be reduced slightly when fish meal is included in the diet. Although the fish meal odor may be objectionable to some animals, the use of a TMR may help to alleviate any serious problems with feed refusals. Although DMI was lower with the diets containing SM, BW changes (Table 5) were not significantly affected. A tendency was observed for slightly higher weight gains on the diets containing SM. This may be a result of less energy going toward the production of
:
:
SBM
Item
22.2' 3.6' Average BW, kg 620 SWChange,kg/wk 1.9 DMI, W d DMI, 46 of BW
SM 2O.Zb 3.3b 613 2.8
USM 21.2&
3.5&
608 3.0
+"Means in same row with unlike superscripts
(P
