Effects of Fish Meals on Rumen Bacterial Fermentation in Continuous Culture1

Effects of Fish Meals on Rumen Bacterial Fermentation in Continuous Culture W. H. HOOVER, T. K. MILLER, and S. R. STOKES

Division of Animal and Veterinary Sciences W. V. THAYNE

Department of Statistics and Computer Science West Virginia University Morgantown 26506

containing fish meal affected microbial metabolism more negatively when the fermentation pH was held at 6.2 than when the pH was 6.0 or lower.

ABSTRACT

Effects of various forms of fish meal on microbial metabolism were investigated in continuous cultures of rumen contents. Five diets were formulated to contain 12% ruminally degradable protein and 47 to 48% nonstructural carbohydrate. Soybean meal was the major protein source in the control diet, whereas in the other four diets, various fish meals were substituted for 6% of total diet DM. Fish meals were: fish meal containing 34.4% FFA, fish meal containing 34.4% F F A with CaC12 added, fish meal containing 65.6% FFA, and fish meal defatted using 1:1 ethanol:ether extraction. The five treatments were fermented with pH either held constant at 6.2 or not controlled. When pH was maintained at 6.2, the inclusion of any fish meal except defatted fish meal reduced the acetate:propionate ratio, decreased protein digestion, and reduced microbial N produced/ per kilogram D M digested when compared with the soybean control. When not controlled, pH decreased after feeding to 6.0 or lower. Under these conditions, the soybean control had a lower acetate:propionate ratio and lower N D F digestion than all diets containing fish meal. In this study, oil-

INTRODUCTION

Received February 13, 1989 Accepted May 1, 1989. ~Published with the approval of the Director of the West Virginia Agricultural and forestry ExperimentStation as Scientific Article Number 2147. Supported by funds provided by the Hatch Act and by Zapata-Haynie Corp., Hammond, LA. 1989 J Dairy Sci 72:2991-2998

The inclusion offish meal (FM) as a source of ruminally undegradable protein in diets for lactating cattle has depressed milk fat percentage (5, 10, 16, 21). Fish meal contains 4.5 to 10.5% fat as determined by the ether extract method (9). Of the total lipids, 14 to 26% are polyunsaturated fatty acids with chain lengths of 20 carbons or more (11). Although microbial lipolytic activity is high, causing rapid formation of F F A in the rumen, the inclusion of unesterified fatty acids in the diet may be more detrimental to rumen microbes than dietary esterified fatty acids (13). In fresh fish oil, fatty acids are components of either neutral lipids or phospholipids, but as the F M deteriorates, F F A are liberated (11); thus, some FM may contain high amounts of FFA. Based on studies primarily with cod liver oil, Storry (18) reported that dietary fatty acids may reduce milk fat by effects on rumen VFA ratios, inhibition of rumen microbial fatty acid synthesis, or -in the case of polyunsaturated fatty acids with 20 or more carbons- reduced uptake of plasma fatty acids by the mammary gland. In order to elucidate the effects of FM on rumen fermentation, a continuous culture study was conducted to determine: 1) microbial metabolic responses to FM when fermentation pH is controlled or not controlled; 2) whether or not F M effects can be modified by infusion of CaC12; 3) the effect of F M

2991

2992

H O O V E R ET AL.

devoid of fat on microbial metabolism; and 4) whether or not the quantity of FFA contained in FM affects the metabolism of rumen microbes. MATERIALS AND METHODS

Diets were prepared containing either no FM, 6% of DM as FM with low FFA (LFFA), or 6% of DM as FM with high FFA (HFFA). In the LFFA FM, 34.4% of the total lipids were free, while the HFFA FM contained 65.6% FFA. The FM and analyses of FFA contents were provided by ZapataHaynie Corp. (Hammond, LA). A fourth diet was formulated containing 6% FM from which the oil was extracted (defatted) using 1:1 ethanol:ether at room temperature. Diet composition and analyses are shown in Table 1. All diets were formulated to provide a similar quantity of degradable protein (12% of DM) and nonstructural carbohydrate. In addition to the control and three FMcontaining diets, a fifth treatment was included in which a CaC12 solution was infused continuously into fermentations of the LFFA diet in an a~tempt to form Ca soaps of the FFA. The infusion was adjusted to provide .32 g CAC12/24 h. The five treatements were fermented in triplicate under two pH conditions: 1) with the pH of the fermentations held constant at 6.20 4- .05, using pH controllers as described by Crawford et al. (2) and 2) with pit uncontrolled and permitted to decrease after feeding. The diets were fed at 100 g D M / d in two equal portions. The continuous culture system was that described by Hoover et al. (8) and modified by Crawford et al. (2). Buffer (20) containing .2 g urea/L was infused to provide a liquid dilution rate of. 12 h i. Solids retention time was 22 h. Incubations consisted of a 6-d equilibration period followed by 3 d of sampling. The composited effluent samples were homogenized 5 min using a SD-45 disperser fitted with a G-450 generator (Tekmar Co., Cincinatti, OH). Effluent DM was determined by centrifuging a weighed sample at 15,000 X g for 45 rain and drying the precipitate residue at 102°C. Journal of Dairy Science Vol. 72, No. 11, 1989

T A B L E 1. Diet composition and analyses. Diets Ingredients Corn silage Grass hay Ground corn Soybean meal Fish meal, high FFA Fish meal, low FFA Fish meal, defatted Urea Mg Sulfate Ca Sulfate DiCal Limestone

High Control F F A 21.5 21.5 35.3 21.1

(% of DM), 21.0 21.0 21.0 21.0 40.6 40.7 10.0 10.0

...

6.0

...... . . . ... .. fl0 .10 .55

Analyses NEt, M c a l / k g 1.69 Crude protein, % 19.2 Soluble protein ~ 16.0 Degradable protein I 64.9 Undegradable protein ~ 35.1 NDF, % 24.6 NSC, 2 % 47.9 Fat, % 3.8 Ca, % .53 P, % .37 Mg, % .22 S, % .23

Low FFA

. . ' i.'10 1.10 .25 .25 . . . . . . . . . . . . . . . . . . . . . . . . . . .

1.68

21.2 21.2 40.4 10.0

...... 6.0

.

Defatted

1.68

... 5.71 1.10 .23

1.68

20.2 27.7

21.5 28.3

20.1 28.3

58.8

58.9

58.8

41.2 25.8 47.8 4.2 .58 .47 .21 .22

41.1 25.2 47.8 4.2 .58 .47 .21 .22

41.2 26.4 47.8 3.7 .56 .46 .21 .21

~Percentage of crude protein. 2Nonstructural carbohydrate.

Although this omits soluble components, mainly VFA, the error is less than that caused by variable amounts of residual buffer salts. Feed DM was determined at 102°C. Feed and effluent NDF and ADF were determined by methods of Goering and Van Soest (6), with modifications of Robertson and Van Soest (14). The procedure of adapting this method to fluid effluent samples has been described (2). Total N in feed, effluent, and microbial samples and ammonia N in effluents were determined by AOAC methods (1) using automated equipment (Tecator Inc., Herndon, VA). The procedure of Webster (19) was used to analyze diaminopimelic acid in mi-

FERMENTATION OF FISH MEALS crobe and effluent samples. Volatile fatty acids were determined by gas chromatograph using a 180-cm X 2-mm i.d. column packed with 10% SP 1200 plus 1% H3PO 4 on 80/100 mesh chromosorb WAW (Supelco, Inc., Bellefonie, PA). Detection was by flame ionization. Lipids in all feeds were determined gravimetrically following extraction with etherethanol (l:l). The FM lipids also were extracted using ethyl ether, petroleum ether, chloroform-methanol (2:1), and hexane. Analyses were conducted using Soxtec equipment (Tecator Inc., Herndon, VA) and methods (3). Nonstructural carbohydrate (NSC) was calculated as: 100 - [percentage CP + (percentage N D F - N D F bound CP) + percentage ether-ethanol extract + percentage ash]. Fermentations were in a completely randomized design with two pH and five diets in a factorial arrangement. Data were analyzed by analysis of variance, and orthogonal comparisons were used to compare diets for each pH. Comparisons used were: 1) low F F A versus high F F A diets, 2) the F M diet with infused CaC12 versus the low FFA and high FFA diets, 3) the defatted F M diets versus the unaltered FM diets, and 4) the control diet versus the F M diets. RESULTS AND DISCUSSION

Analyses of FM total lipids using various solvents are in Table 2. These data demonstrate the importance of solvent selection when attempting to quantitate total lipids in feeds of this type. Although chloroform-

2993

methanol was the most effective solvent, ether-ethanol was used to defat the F M for diet preparation to minimize protein denaturation. For consistency, ether-ethanol was subsequently used for all diet analyses. Digestion coefficients are presented in Table 3. When not controlled, the pH decreased to an average of 6.0 or below for all treatments. This caused expected decreases in all digestion coefficients for the control diet, but D M digestion was the only parameter that did not have a pH X treatment interaction. In light of this, treatment results will be discussed for each pH separately. At pH 6.2, the major effects of F M on nutrient digestibilities were the significant increases in D M and protein digestion with the diet containing defatted F M compared with oil-containing F M diets. This is clearly seen in the orthogonal contrast of the defatted diet results with the mean of the other FM diets, where the defatted diet was significantly greater in D M and protein digestion. At pH 6.2, infusion of CaC12 was effective in partially removing the negative influence of the fish oil, with orthogonal contrasts showing significantly greater DM and protein digestion in the F M diet with CaC12. No differences in DM, NDF, or A D F digestion were noted when the soybean control diet was compared to the diets with FM, suggesting the presence of oil-containing F M did not affect digestion of nonprotein feed components. This agrees with the other findings (15, 21), that reported no depressions in organic matter digestion when F M was added to the diets of cattle.

TABLE 2. Lipid content of high free fatty acid (HFFA), low free fatty acid (LFFA), and defatted fish meals using various solvents. Solvent Chloroform: Ethyl ether: Fish meal Ethyl ether Pet ether methanol Hexane ethanol HFFA LFFA Defatted~

6.1 6.9 NA2

6.1 8.0 NA

(% of DM) 11.3 12.2 NA

5.9 8.0 NA

9.7 9.7 1.2

~Defatted by soaking 1 kg fish meal in 8 L of ethyl ether:ethanol(1:1) for sevenconsecutive24-h periods. Solvent decanted and replaced each period. 2Not analyzed. Journal of Dairy ScienceVol. 72, No. 11, 1989

2994

HOOVER ET AL.

TABLE 3, Effects of fish meal on nutrient digestibilities in continuous culture at two pH conditions. ~ Diet2 Digestion (%) DM a NDF b ADFb Protein b

pH

Control

Def.3

Low FFA + CaC124

6.2d,e UC 6.2 UCf 6.2 UCc,f 6.2dx UCJ

71.8 60.7 40.4 14.7 33.9 15.6 68.4 50.7

81.6 69.7 48.6 33.7 38.5 21.1 83.0 38.5

79.4 60.2 44.6 29.9 38.3 27.6 62.8 58.1

FFA Low

High

65.7 64.9 45.5 32.2 31.1 26.8 43.0 42.9

72.0 63.3 42.1 37.3 34.5 33.1 56.2 44.4

apH effect (P<.05). bpH )< treatment interaction (P<.05). cLow vs. high FFA (P<.05) at specified pH. dlnfusion of CaC12vs. LFFA and HFFA diets (P<.05) at specified pH. ~Defatted vs. other fish meals (P<.05) at specified pH. fControl vs. fish meals (P<.05) at specified pH. JThe pH was held at 6.2 or not controlled (UC). Whe not controlled, mean pH for control, defatted, CaCI2, low FFA, and high FFA diets were 5.8, 5.6, 6.0, 5.8, and 5.9, respectively. 2Least square means averaged across three fermentations per treatment. 3Defatted fish meal. 4The CaCI~ was infused into fermenters at ,32 g/24 h. Total protein digestion was 68% on the soybean meal diet and had a m e a n of 50% for the H F F A and L F F A diets. This is in agreement with Zerbini et al. (21), who reported ruminal protein degradability values of 66 and 47% for soybean and F M diets, respectively, when fed to lactating cows. R o o k e and A r m s t r o n g (15), however, reported a decrease in total protein digestion of only 10 percentage units u p o n addition of F M to barleysilage diets. T h e presence of high a m o u n t s of F F A , which is indicative of a more deteriorated product, did not affect nutrient digestibilities to any greater extent than did the diet with L F F A . W h e n p H was not controlled, all mean fermentation p H were between 5.8 and 6.0 except for the deflated treatement, which was lower (P<.05) at 5.6 (Table 3). Under these conditions, nutrient digestion responses to F M were considerably different than at p H 6.2. Digestion of A D F and N D F was improved by F M c o m p a r e d with that of soybean meal control. Protein digestion was not improved by defatting the F M , but it was greater Journal of Dairy Science Vol. 72, No. 1l, 1989

when CaC12 was infused; the latter response was also noted at p H 6.2. Total V F A p r o d u c t i o n and m o l a r ratios in both the pH-controlled and uncontrolled fermentations are in Table 4. At p H 6.2, all diets with oil-containing F M tended to have lower acetate and had significantly higher p r o p i o n ate m o l a r percentages than diets with defatted F M . This resulted in a significantly lower acetate:propionate ratio than for the contol or defatted F M diets. The changes in acetate:propionate ratio, along with a tendency for less butyrate in the oil-containing diets, is consistent with r u m e n conditions associated with reduced milk fat production. Thus, observed depressions in milk fat p r o d u c t i o n (5, 10, 16, 21) may be associated with changes in the r u m e n as well as decreased uptake of plasma fatty acids at the m a m m a r y gland as suggested by Storry (18). Similar acetate:propionate ratio changes due to F M were reported by R o o k e and A r m s t r o n g (15), and, although Zerbini et al. (21) reported a significant decrease in acetate m o l a r percentage on a F M diet, no change in propionate occurred.

2995

FERMENTATION OF FISH MEALS

TABLE 4. Effects of fish meal on volatile fatty acid production in continuous culture at two pH conditions.1 Diet2 FFA

Parameter

pH

Control

Def.3

Low FFA ÷ CaCI24

Total VFA, M/24 ha

6.2a UC

.41 .34

.40 .36

.44 .36

.41 .35

.44 .35

Acetate,b mol/100 mol

6.2 UC

54.7 46.9

56.4 49.3

52.0 48.8

51.4 46.5

51.1 50.8

Propionate, b mol/I00 tool

6.2e UCc,f 6.2 UCf 6.2 UCe,f 6.2 UCe 6.2 UCc 6.2e,f UCc,f

21.2 35.0 18.9 12.5 2.2 4.4 .7 .2 2.4 1.2 2.60 1.40

19.2 30.2 18.6 14.5 1.9 4.8 .7 .1 3.3 1.1 2.94 1.66

26.6 25.3 16.3 20.9 2.1 2.2 .6 .4 2.5 2.4 1.97 1.94

26.4 30.5 17.7 17.9 1.9 2.8 .5 .2 2. i 2.1 1.95 1.55

26.5 23.1 17.3 20.9 1.9 2.7 .6 .4 2.7 2.1 1.93 2.21

Butyrate, mol/100 mol Valerate,b mol/100 mol Isobutyrate,b mol/100 tool Isovalerate,a mol/100 mol A:Pb

Low

High

apH effect (P<.05). bpH X treatment interaction (P<.05). cLow vs. high FFA (P
to high grain diets that caused a r e d u c t i o n in r u m e n pH. This type of response was seen in the studies of H a s s a n and B r y a n t (7), who f o u n d that i n c l u s i o n of F M in a 60:40 forage to g r a i n diet reduced the a c e t a t e : p r o p i o n a t e ratio to 3.5 f r o m a c o n t r o l value of 4.2. W h e n the diet was switched to a 40:60 forage to g r a i n ratio, the a c e t a t e : p r o p i o n a t e ratio was 2.5 for the c o n t r o l and was similar at 2.4 for the F M diet.Also, in studies with lactating cows (10), significant milk fat d e p r e s s i o n resulted when F M was added to the diets of cows c o n s u m i n g 51% of diet D M as c o n c e n trate. W h e n the cows were fed diets with 72% concentrate, milk fat was low (2.47%) o n the c o n t r o l diet b u t was significantly increased to 2.74 u p o n i n c l u s i o n of 12.1% F M in the diet. Journal of Dairy Science Wol. 72, No. 11, 1989

2996

HOOVER ET AL.

TABLE 5. Effects of fish meal on nitrogen partitioning in continuous culture at two pH conditions3 Diet2 FFA Parameter

pH

Ammonia N, b mg/dl

6.2d,f UCf

Nonammonia N, b g.d

Control

Def. 3

CaC124

Low

High

8.2 1.0

14.5 3.2

17.8 6.9

13.7 4.5

9.9 7.9

6.2e UC c

3.1 3.6

3.0 3.4

3.1 3.5

3.3 3.6

3.2 3.3

Microbial N, b g/d

6.2c,e UCd

2.1 1.8

2.5 1.5

1.9 2.1

1.3 1.6

1.8 1.5

Undegraded feed N,b g/d

6.2°,d,e UC d

1.0 1.5

.6 2.0

1.3 1.4

2.0 2.0

1.4 1.8

Undegraded intake protein, b %

6.2d,e UCd

31.6 49.3

17.0 61.5

37.2 41.9

57.0 57.1

43.8 55.6

Microbial N/kg DM digested b

6.2c UCdx

29.5 30.5

30.6 20.7

23.4 34.2

20.0 25.1

24.9 23.5

apH effect (P<.05). bpH X treatment interaction (P<.05). cLow vs. high FFA (P<.05) at specified pH. qnfusion of CaC12 vs. LFFA and HFFA diets (P<.05) at specified pH. eDefatted vs. other fish meals (P<.05) at specified pH. fControl vs. fish meals (P<.05) at specified pH. ~The pH was held at 6.2 or not controlled (UC). When not controlled, mean pH for control, defatted, CaC12, low FFA, and high FFA diets were 5.8, 5.6, 6.0, 5.8, and 5.9, respectively. 2Least square means averaged across three fermentations per treatment. 3Defatted fish meal. 4The CaC12 was infused into fermenters at .32 g/24 h. O r s k o v et al. (12), however, r e p o r t e d no effect on m i l k fat w h e n up to 12% F M was a d d e d to the diets o f l a c t a t i n g cows. The n i t r o g e n p a r t i t i o n i n g d a t a are in T a b l e 5. A g a i n , t h e r e were significant p H X t r e a t m e n t interactions. A t p H 6.2, the effects o f the fish oil are e v i d e n t in the o r t h o g o n a l c o n t r a s t s o f the d e f a t t e d vs. the o t h e r F M t r e a t m e n t s . The diets with the o i l - c o n t a i n i n g meals h a d m o r e u n d e g r a d e d feed N a n d less m i c r o b i a l p r o t e i n synthesis t h a n in diets with d e f a t t e d F M . M i c r o b i a l efficiency ( g r a m s o f m i c r o b i a l N / p e r k i l o g r a m D M digested) also was sign i f i c a n t l y h i g h e r for the diet with d e f a t t e d F M . Because all diets were f o r m u l a t e d to p r o vide a d e q u a t e d e g r a d a b l e N a n d N S C to supp o r t e q u a l m i c r o b i a l g r o w t h across t r e a t ments, the lower m i c r o b i a l efficiencies a p p a r e n t l y reflect a negative influence of fish oil on m i c r o b i a l m e t a b o l i s m . T h e t o t a l p r o tein of the diet c o n t a i n i n g H F F A was m o r e Journal of Dairy Science Vol. 72, No. 11, 1989

d e g r a d a b l e and s u p p o r t e d g r e a t e r m i c r o b i a l N p r o d u c t i o n t h a n the diet with L F F A . T h e a d d i t i o n o f CaC12 to the L F F A diet increased the p r o t e i n d e g r a d a b i l i t y . W h e n p H was n o t c o n t r o l l e d , p r o t e i n d e g r a d a b i l i t y o f the c o n t r o l r a t i o n was l o w e r t h a n at p H 6.2, b u t efficiency o f m i c r o b i a l g r o w t h r e m a i n e d similar to t h a t at p H 6.2. W h e n d e f a t t e d m e a l was fed, m i c r o b i a l efficiency was d e c r e a s e d relative to t h a t on the o t h e r F M diets. T h e r e a s o n for the d e c r e a s e d m i c r o b i a l g r o w t h is n o t clear o t h e r t h a n t h a t the p H o f the f e r m e n t a t i o n (average 5.6) was significantly lower t h a n t h a t of all o t h e r ferm e n t a t i o n s . W i t h i n the o i l - c o n t a i n i n g meals, the a d d i t i o n of CaC12 resulted in increased protein degradability, microbial growth, and m i c r o b i a l efficiency. S i m i l a r changes in N m e t a b o l i s m due to CaC12 were n o t e d at p H 6.2 and m a y be related to d e c r e a s e d t o x i c i t y of the s o a p s vs. the F F A .

2997

FERMENTATION OF FISH MEALS

Overall, the rumen microbial responses to the inclusion of FM in the diet observed in this study were similar to results of others. Zerbini et al. (21) noted decreased microbial N production and efficiency when F M was fed. In their study the rumen pH on the FM diet was 6.3, and our results at pH 6.2 show similar responses. Rooke and Armstrong (15) reported results of including FM into diets in which rumen pH was maintained at 6.7 to 6.8. In their study, microbial N tended to be lower on the F M diets, as in the current study, but microbial efficiency remained unchanged. Dawson et al. (4) found significantly increases in both total microbial N flow and microbial efficiency when FM was added to an all-silage diet. No values for FM effects at low rumen pH were found in the literature. CONCLUSIONS

Effects of F M addition to the diet varied with pH. When pH was maintained at 6.2, digestion of carbohydrates was not affected by FM, but the acetate:propionate ratio was greatly reduced and microbial protein production and efficiency were impaired. These effects were not noted when the diet was prepared with defatted FM, indicating that the effects were due to the fatty acid content. Infusion of CaCI2, in an attempt to form soaps, increased DM and protein digestion but did not affect VFA ratios or microbial growth. When the fermentation pH was permited to cycle, the mean pH of the treatments varied from 5.6 to 6.0. Under these conditions, fiber digestion was greatly reduced on the soybean meal control diet but improved by addition of either the oil-containing or defatted FM. The FM also decreased propionate and increased the acetate:propionate ratio relative to the soybean meal control diet. When pH was uncontrolled, CaC12 infusion improved microbial growth and efficiency of the L F F A diet compared with the same diet without infusion. When compared with the diet with F M of low F F A content, the use of a lower quality product containing high F F A had minimal effects on microbial metabolism. Although protein degradation was higher for the high

F F A FM, the N was efficiently converted to microbial protein rather than to ammonia. From these data it appears that under feeding conditions in which rumen pH is maintained at 6.2 or above, use of F M would cause shifts in rumen VFA ratios normally associated with milk fat depression. When rumen conditions were below pH 6.0, dietary F M improved fiber digestion and VFA ratios, suggesting its use would not exacerbate ruminal factors that would be conducive to low milk fat percentage. REFERENCES 1 Association of Official Agricultural Chemists. 1984. Official methods of analysis. 14th ed. Assoc. Offic. Agric. Chem., Washington, DC. 2 Crawford, R. J. Jr., B.J. Shriver, G. A. Varga, and W. H. Hoover. 1983. Buffer requirements for maintenance of pH during fermentation of individual feeds in continuous cultures. J. Dairy Sci. 66:1881. 3 Czechoslovakia Institute of Agriculture. 1986. The determination of crude fat in feed using the soxtec system HT. Focus, Tecator J. Technoh Chem. Anal. 9:6. 4 Dawson, J. M., C. I. Bruce, P. J. Buttery, M. Gill, and D. E. Beever. 1988. Protein metabolism in the rumen of silage-fed steers - effect of fish-meal supplementation. Br. J. Nutr. 60:339. 5 Erfle, J. D., S. Mahadevan, R. M. Teather, and F. D. Sauer. 1983. The performance of lactating cows fed urea-treated corn silage in combination with soybean or fish meal containing concentrates. Can. J. Anim. Sci. 63:191. 6 Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analysis. Agric. Handbook No. 379. ARS, USDA, Washington, DC. 7 Hassan, S. A., and M. J. Bryant. 1986. The response of store lambs to dietary supplements of fish meal. I. Effects of forage-to-concentrate ratio. Anim. Prod. 42:223. 8 Hoover, W. H., B. A. Crooker, and C. J. Sniffen. 1976. Effects of differential solid-liquid removal rates on protoza numbers in continuous culture of rumen contents. J. Anim. Sci. 43:528. 9 National Research Council. 1988. Nutrient requirements of dairy cattle. 6th ed. Natl. Acad. Sci., Washington, DC. 10 Oldham, J. D., D. J. Napper, T. Smith, and R. J. Fulford. 1985. Performance of dairy cows offered isonitrogenous diets containing urea or fish meal in early and mid-lactation. Br. J. Nutr. 53:337. 11 Opstvedt, J. 1985. Fish lipids in animal nutrition. Tech. Bull. No. 22. Int. Assoc. Fish Meal Manuf., Hertfordshire, UK. 12 Orskov, E. R., G. W. Reid, and C. A. G. Tait. 1987. Effect offish meal on the mobilization of body energy in dairy cows. Anim. Prod. 45:1987. 13 Palmquist, D. L., and T. C. Jenkins. 1980. Fat in Journal of Dairy Science Voh 72, No. 11, 1989

2998

HOOVER ET AL.

lactation rations:review. J. Dairy Sci. 63:1. 14 Robertson, J. B., and P. J. Van Soest. 1977. Dietary fiber estimation in concentrate feedstuffs. J. Anita. Sci. (Suppl. 2) 45:254. 15 Rooke, J. A., and D. G. Armstrong. 1987. The digestion by cattle of silage and barley diets containing increasing quantities of fish meal. J. Agric. Sci. Camb. 109:261. 16 Sloan, B. K., P. Rowlison, and D. G. Armstrong. 1988. The influence of a formulated excess of rumen degradable protein or undegradable protein on milk production in dairy cows in early lactation. Anita. Prod. 46:13. 17 Statistical Analysis Systems. 1982. SAS User's guide. SAS Inst., Cary, NC.

Journal of Dairy Science Vol. 72, No. 11, 1989

18 Storry, T. E. 1981. The effect of dietary fat on milk composition. Page 3 in Recent advances in animal nutrition. W. Haresign, ed. Butterworths, London, UK. 19 Webster, P. M. 1986. Effect of various forms of fat on rumen fermentation in continuous culture at varying pH. M. S. Thesis, West Virginia Univ., Morgantown. 20 Weller, R. A., and A. F. Pilgrim. 1974. Passage of protozoa and volatile fatty acids from the rumen of the sheep and from a continuous in vitro fermentation system. Br. J. Nutr. 32:341. 21 Zerbini, E., C. E. Polan, and J. H. Herbein. 1988. Effect of dietary soybean meal and fishmeal on protein digesta flow in Holstein cows during early and mid-lactation. J. Dairy Sci. 71:1248.