Maataloustieteellinen A ikakauskirja
Vol. 59: 101—120, 1987

The effect of dietary inclusion of barley, unmolassed sugar beet pulp and
molasses on milk production, digestibility and digesta passage
in dairy cows given silage based diet

PEKKA HUHTANEN

Department of Animal Husbandry, University of Helsinki
SF-00710 Helsinki, Finland

Abstract. Ten dairy cows in early lactation were given five different supplements with grass
silage ad libitum according to two 5x5 Latin square designs. The supplements consisted of
barley (Ba), unmolassed sugar beet pulp (SBP) and molasses (Mo) on dry matter (DM) basis
as follows: (A) Ba 1000 g, (B) Ba 667 g and Mo 326 g, (C) Ba 417 g, SBP 410 g and Mo 163
g, (D) SBP 983 g and (E) 656 g SBP and 326 g Mo per kg. The lower fat content of SBP and
molasses than of Ba was balanced with rapeseed oil. These supplements were fed at the level
of 6 kg DM/d. In addition, 1 kg of rapeseed meal was given as a protein supplement.

No differences in silage DM intake between the Ba and SBP supplements were found,
but inclusion of a high level of Mo depressed silage and total DM intake. Cows given SBP
supplements yielded 1.1 kg (P < 0.01) more milk and 32 g (P < 0.05) more protein than those
given Ba supplements. Milk fat concentration was lower (P < 0.001) on SBP (46.0 g/kg) than
on Ba diets (49.1 g/kg) and thus there was no difference in the yield of fat corrected milk (FCM)
and milk fat between the two supplements. Inclusion of 2 kg of DM of Mo reduced the milk
(P < 0.001), FCM, fat and protein yields (P < 0.01) compared to diets without molasses.
Supplement C increased (P < 0.05) the milk yield compared to the other diets but the milk
composition was not affected.

The effect of supplements on the digestibility of the ration was determined using acid
insoluble ash as a marker. The apparent digestibility of organic matter averaged 0.743 and
was not significantly affected by the diet. Digestibility of neutral detergent fibre was higher
(P < 0.001) for SBP diets (0.680) than for Ba diets (0.596). Similar differences were observed
in the digestibility of other fibre components. However, the digestibility of the silage was not
affected by the treatment. Feeding SBP diets resulted in lower (P < 0.001) digestibility of crude
protein than did Ba diets. Metabolizable energy (ME) of SBP diets tended to be used more
efficiently for milk synthesis than ME of Ba diets. Inclusion of Mo in Ba or SBP did not reduce
the efficiency, although the milk yield was decreased.

The passage rate of liquid was determined with CoEDTA and that of particles with
Cr-labelled straw. The average mean retention time of liquid ranged from 18.0 to 19.9 h and
that of particles from 36.8 to 37.7 h. Neither time was affected by the treatment. The passage
rate of particles from the rumen ranged from 0.045 to 0.048 and that of liquid from 0.096
to 0.104, irrespective of the diet.

Index words: milk production, barley, unmolassed sugar beet pulp, molasses, silage, digestibility, digesta passage

101

JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND

https://www.c-info.fi/en/info/?token=feU3agR5dHv5yNWR.I8jgKDHO343iurFbaartyA.2JIwNRapk9Qbk18QL9wHewtcIfZrfC5a6h2pEqwkVRTZYwqMsi5SIojZm_tofel63fcirboKHs86yem9v3BwxgUtH7mbptQiuEtH2sqzdBAZ3RMMmTuyQuycD_knUBSxMdD3JuEDytmrYci4p3eoLRISMUOzyjOnt24EHg


Introduction

For high producing dairy cows even the best
quality forage fed ad libitum provides in-
sufficient nutrients. To meet the energy and
protein requirements, supplementary feeding
is needed. Energy supplements are primarily
based on cereal grains, and protein supple-
ments on oilseed meal.

Feeding of concentrates withhigh sugar and
starch contents at high levels of supplemen-
tation has increased the number of digestive
disturbances (De VISSER and De Groot 1981)
and lowered the concentration of milk fat
(Sutton et al. 1985), but only when the sup-
plement comprised a large proportion of the
diet. Furthermore, increasing the ratio of
grain to forage has changed the energy parti-
tion towards body fat deposition (Flatt et al.
1969). Addition of starch to forage based diets
leads to a reduction in cellulolytic activity in
the rumen (El-Shazly et al. 1961, Mertens
and Loften 1980). Reduction in the rate and
extent of forage digestion, slowing the passage
of digesta, will result in decreased forage in-
take. Replacing starchy supplements by fi-
brous supplements has widely been reported
to increase the digestibility of different fibre
components (Mayne and Gordon 1984,
Sutton et al. 1985, Thomas et al. 1986).
Fahmy et al. (1984) reported a smaller de-
crease in the disappearance of ammonia-
treated straw from nylon bags in sheep re-
ceiving sugar beet pulp than in sheep receiving
barley with increasing levels of supplementa-
tion. Similarly, Istasse et al. (1986) reported
a higher digestibility of ammonia-treated
straw in dairy cows on sugar beet pulp based
diet than on barley based diet.

At present, sugar beet pulp and wheat bran
are the most important fibrous by-products
used in the feeding of dairy cows in Finland.
In the Finnish feed tables (SALOet al. 1982),
it is suggested that 1 kg of dry matter of sugar
beet pulp is equivalent to 0.87—0.90 kg of dry
matter of barley. However, replacing barley
with SBP on equal dry matter basis in differ-
ent forage diets has resulted in similar or

slightly better productive value in milk pro-
duction (Castle 1972, Castle et al. 1981,
Mayne and Gordon 1984, Murphy 1985,
Thomas et al. 1986, Istasse et al. 1986).

The purpose of the present study was to
evaluate the comparative feeding value for
milk production of rolled barley, unmolassed
sugar beet pulp and molasses when used as
supplements with a diet of grass silage ad
libitum. In addition, the effects of the sup-
plements on the feed intake, the digestibility
of the ration and the passage rate were deter-
mined. The effect of the supplements on
certain blood metabolites and hormones will
be reported later in detail (Miettinen and
Huhtanen 1987).

Materials and methods

Animals and management

Ten autumn calving Ayrshire cows were
used in the experiments. Five of the animals
were heifers, and five were cows in their third
or fourth lactation. On average, the cows had
calved 41 (SE 3) days before the start of the
experiment. The average fat corrected milk
(FCM) yield of the cows during the previous
lactation was 7915 (SE 96) kg. The cows were
fed and housed individually. They were fed
at 5 and 14 hours and given access to food for
3 h per feeding time. Refusals were removed
and weighed after the morning feeding.

Design and treatments

The experimental design was two 5x5
Latin squares, vMth sqdares balanced so that
every treatment'was (wice preceded by each
of the other treatments. Five different sup-
plements (Tqblevl) were given with a grass
silage diet. Energy supplements were given on
a flat rate basis of 6 kg dry matter (DM) daily
throughout the experiment. The daily al-
lowance of concentrate was adjusted to supply
the same amount of DM. The lower content
of ether extract in the unmolassed sugar beet
pulp (SBP) and sugarbeet molasses (Mo) than

102



Table I. Composition of experimental diets.

Treatment
A B C D E

Grass silage ad lib ad lib ad lib ad lib ad lib
Rapeseed meal (kg/d) 1.0 1.0 1.0 1.0 1.0
Concentrate (kg DM)

Barley 6.0 4.0 2.48
Beet pulp 2.48 5.9 3.93
Beet molasses 1.96 0.98 1.96
Rapeseed oil 0.04 0.06 0.10 0.11

Minerals' (kg/d) 0.25 0.25 0.25 0.25 0.25

In mineral mixture (g/kg): Ca 170, P 80, Na 60, Mg 80.

in barley (Ba) was compensated by the addi-
tion of rapeseed oil. Barley was fed in rolled
form. The ingredients of the concentrate were
weighed separately and mixed before feeding.
Direct cut silage was made from secondary
growth of timothy, meadow fescue and red
clover sward (45:35:20). The crops were har-
vested with a precision chop forage harvester
and ensiled with formic acid additive (5 1/t).
Silage was given ad libitum in such a way that
refusals accounted for about 10 % of the
amount offered.

Each experimental period lasted 21 days.
The changes in the diet were made within the
first 4 days of the period. Except for live
weight and live weight change, the results are
based on the data of the third week of the
period.

Milk yield was recorded at every milking.
Milk samples were taken on days 10, 17 and
18 of each period and each set was combined,
in proportion to yield, for analyses. The cows
were weighed at the beginning of the experi-
ment and on days 6,7, 20 and 21 of each
period. Feed samples were taken weekly and
analysed for DM. For other analyses the sub-
samples of three weeks were pooled.

Digestibility measurements

Apparent digestibility of the diets was de-
termined in the five older cows using acid in-
soluble ash (AIA) as a natural marker (Van
Keulen and Young 1977). Faecal grab sam-

pies were taken on 6, 14 and 22 hours for 5
days from day 15 onwards. Pooled samples
of faeces were dried at 100°C for 48 h and
milled to pass through a 1-mm screen.

Digesta passage measurements

Two markers were used to measure digesta
passage in the five older cows: liquid passage
was determined with CoEDTA and the pas-
sage of particles with Cr-labelled straw. The
markers were prepared according to Uden et
al. (1980). A single dose of 20 g of CoEDTA
and 100 g of Cr-straw mixed with concentrate
was given on day 15 of each period at the
morning feeding. Faecal grab samples were
taken at 4 (8 —36 h), 6 (36 —48 h) and 8 h
(48 —112 h) intervals. Samples were dried at
100°C for 48 h and milled to pass through a
1-mm screen.

The rate constants for passage of Cr were
estimated according to the equation of
Growum and Williams (1973), assuming a
two pool model:

y = Aerk t,—Tr>— Ae~kA'~ TT) for t > TT
y = 0 for t < TT (1)

where y and A are the marker concentrations
in faecal DM, k, and k 2 are the rate con-
stants associated with the kinetics of the mark-
er in the reticulorumen and hindgut and t is
the sampling time (h). TT, transit time (h), is
the estimated time for the first appearance of
marker in the faeces:

103



_

InA 2 InA, (2)
k 2 —k,

where A, and A 2 are intercept values for the
marker concentration in faecal DM and k,
and k 2 are as defined above. Values on the
linear declining portion of the line were de-
termined by eye. Because the maximum con-
centration of Co was found in either the first
or second sample after dosing of the marker,
it was not possible to estimate k 2 for the
liquid passage rate. Therefore the constant for
the passage of liquid was calculated assuming
a one-pool model (Grovum and Williams
1973):

c = (3)

where C is the concentration of Co in faeces
at time t and C 0 the concentration at time tO,

and t is the sampling time (h).
Total mean retention time (TMRT) was cal-

culated according to Thielemans et al.
(1978):

ZfpAt,
TMRT= (4)

EC,A/,

where t is the time (h) from dosage of the
marker to the mid-point of the i'h collection
interval, C( is the concentration of the mark-
er in the i'h sample, At, is the interval between
two samplings and n is the number of faecal
samples. TMRT was divided into pool mean
retention time (PMRT) and residual mean
retention time (RMRT). PMRT was calcu-
lated as \/k, and RMRT as TMRT-PMRT.
PMRT was assumed to be the pool effect of
rumen and RMRT the effect of caecum and
colon. TMRT was also calculated according
to Grovum and Phillips (1973).

TMRT = TT + 1/k, + 1 /k2 (5)

Energy utilization
The energy requirement for maintenance

was calculated from equations of the Agricul-

tural Research Council (ARC 1980) and the
requirement for live weight change according
to the Ministry of Agriculture, Food and Fish-
eries (Maff 1975). The gross energy of milk
was calculated with the equations of Tyrrel

and Reid (1965). Calculation of digestible
organic matter (DOM) intake of cows and
heifers was based on the results of the digest-
ibility trial. DOM was converted to digestible
energy (DE) assuming an energy content of 19
MJ/kg DOM (ARC 1980), and DE was con-
verted to metabolisable energy (ME) using a
factor of 0.86.

Analytical methods

The DM content of feeds was determined
by oven drying at 103°C for 24 h. Before anal-
ysis the samples were dried in vacuum and
milled through a 1-mm screen. Feed analyses
were made according to standard procedures.
Silage DM content was corrected for volatile
losses of lactic acid, volatile fatty acids (VFA)
and ammonia as described by Porter et al.
(1984). Ammonia N (McCullough 1967),
lactic acid (Barker and Summerson 1941),
sugars (Nelson 1944) and VFA (Huida 1973)
were analysed and pH measured of fresh silage
samples. Neutral detergent fibre (NDF), acid
detergent fibre (ADF) and acid detergent
lignin (ADL) were analysed according to
Goering and Van Soest (1970), but as mod-
ified by Robertson and Van Soest (1977)
for barley. Starch content was analysed by the
method of Salo and Salmi (1968). Net
energy in fattening feed units (FFU) and di-
gestible crude protein (DCP) were calculated
according to Finnish feed tables (Salo et ai.
1982), and ME values according to MAFF
(1975). The chemical composition and feed
values of the experimental feeds are presented
in Table 2.

Milk fat and protein contents were analysed
with an infrared milk analyser (IRMA), and
the fatty acid composition of milk was deter-
mined by gas chromatography.

AIA was determined in feed and faecal
samples using 2 N HCI, as described by Van

104



105

Table 2. Chemical composition of experimental feeds (g/kg DM) and feed values.

Silage Barley Beet pulp Molasses Rapeseed meal

Dry matter (g/kg) 227 867 872 773 878
In dry matter

Ash 90 27 53 147 77
Crude protein 174 118 124 148 341
Ether extract 58 20 8 nd. 76
Crude fibre 311 65 210 nd. 145
NEE 1 367 769 605 705 361
NDF 533 213 510 nd. 270
ADF 375 71 266 nd. 223
ADL 31 13 25 nd. 100
Cellulose 345 58 241 nd. 123
Hemicellulose 157 142 243 nd. 47
Sugars 27 37 86 616 87
Starch nd. 563 29 nd. 17

Feed values
FFUVkg DM 0.731 1.139 1.027 0.903 1.011
ME MJ/kg DM 10.32 13.28 11.66 12.56 11.96
DCP’ g/kg DM 122 86 79 104 285

nd. =not determined, 'NFE =nitrogen free extracts, 2FFU =fattening feed unit =0.7 kg starch, 3DCP = digestible
crude protein.
In silage: pH 4.08; in dry matter (g/kg): lactic acid 50, acetic acid 16, propionic acid 0.3, butyric acid 0.2; in total
nitrogen (g/kg): NH, —N 46, soluble N 460; D-value 0.622.

Keulen and Young (1977). Cr and Co were
analysed by atomic absorption spectrophotom-
etry after treating the samples as described by
Williams et al. (1962).

Statistical analysis

Analysis of variance for Latin square ex-
periments was used for digestibility and rate
of passage measurements. The model used for
statistical analysis was

y
Uki = T, + Cj + Pk + eijk\<

where T, C, P are treatment, cow and period
effects. The model used to analyse the data
from the two Latin squares was:

y/jkim = S, + C/SJ + Pk + T, + (SP),k
+ (ST) U + eiJk ,m>

where S, C, P and Tare square, cow, period
and treatment effects and SP and ST square
X period and square x treatment interac-
tions. The treatment effect was further
partitioned, using orthogonal contrasts, into
effects due to differences between barley and

SBP diets (AB v. DE), differences resulting
from replacing Ba or SBP by molasses (AD
v. BE), or differences between treatment C v.
other treatments and interaction between type
of supplement (Ba v. SBP) and level of
molasses (AE v. BD).

Results

Feed intake

The effect of the treatments on feed intake
and nutrient consumption are given in Table
3. Replacing 330 g/kg of barley or SBP with
molasses on DM basis tended to decrease
(P = 0.065) silage DM intake, and the associ-
ated difference in total DM intake was signif-
icant (P < 0.05). Silage intake was similar on
Ba and SBP diets. However, the interaction
between the level of molasses and type of
supplement in DM intake, in addition to FEU,
ME and DCP intakes, was significant (P <

0.05). Some cows on treatment D refused
small amounts of SBP and the concentrate in-
take was slightly lower (P > 0.05) for diet D



Table 3. Feed intake (kg DM/d) and nutrient consumption in cows given grass silage with different energy supple-
ments.

Statistical significance
of effectTreatment SEM

A B C D E 24 df AB v. AD v. AE v. C. v.
DE BE BD other

Grass silage
Barley

9.37 8.51 8.76 9.21 8.97 0.28 NS NS NS NS
5.96 3.87 2.48 0.05

Beet pulp
Molasses

2.43 5.50 3.86 0.09
1.92 0.97 0.07 1.94 0.04

0.87 0.84 0.87 0.83 0.85 0.002Rapeseed meal 0.87 0.84 0.87 0.83 0.85 0.002
Concentrate total 6.83 6.63 6.74 6.40 6.66 0.15 NS NS NS NS
Total DM intake 16.21 15.14 15.50 15.61 15.63 0.22 NS * * NS
DM intake as %

of live weight 3.16 2.97 3.01 3.04 3.06 0.042 NS NS * NS
DM intake as

g/kgW075 150.4 141.0 143.1 144.5 145.2 2.27 NS NS * NS
FFU/d 14.53 13.30 13.59 13.48 13.35 0.17 *** •• NS
ME MJ/d 186.6 174.2 175.4 171.9 174.3 2.34 ** � *• NS
DCP g/d 1899 1784 1805 1807 1828 30.6 NS NS * NS

SEM =standard error of means
Significance: NS (non-significant), *(P<O.O5), **(P<o.ol), �**(P<o.ool).

Table 4. Daily intake (kg) of crude protein, ether extract, different carbohydrates and crude fibre in cows given
grass silage with different energy supplements.

Statistical significance
of effectTreatment

SEM
A B C D E 24 df AB v. AD v. AE v. C. v.

DE BE BD other

Crude protein 2.63 2.50 2.55 2.58 2.61 0.04 NS NS NS NS
Ether extract 0.73 0.63 0.68 0.73 0.68 0.01 NS *** NS NS
Starch 3.37 2.19 1.48 0.17 0.13 0.03 *** *»* NS NS
Sugars 0.55 1.65 1.21 0.85 1.85 0.03 *** »** NS NS
Cellulose 3.68 3.26 3.89 4.58 4.13 0.08 *** *** NS NS
Hemicellulose 2.36 1.93 2.36 2.81 2.39 0.04 •** *** NS NS
Crude fibre 3.42 3.02 3.53 4.10 3.72 0.07 •** *•* NS NS

For significance: see Table 3.

than for other treatments. Estimated FFU and
ME intakes were higher (P < 0.01) in cows
given Ba diets than in those given SBP diets.
Inclusion of 2 kg of molasses (on DM basis)
in the diet resulted in lower FFU (P<0.001)
and ME intake (P < 0.05) compared to diets
without molasses.

The different carbohydrate composition of
concentrate ingredients caused marked differ-
ences in the intake of the various carbohy-
drates (Table 4). Ba and SBP diets differed

in composition primarily in terms of the ratio
of cell wall carbohydrates to starch. When
molasses was included in the Ba diet, sugars
were substituted for starch; and in the SBP
diet they were substituted for cellulose and
hemicellulose.

Milk yield and milk composition
Dietary inclusion of 2 kg of molasses re-

duced significantly (P < 0.001; P < 0.01)
the milk, FCM, milk fat and protein yields

106



Table 5. Milk yield, milk composition and feed conversion in cows given grass silage with different energy supplements.

Statistical significance
of effectTreatment SEM

A B C D E 24 df AB v. AD v. AE v. C. v.
DE BE BD other

Milk yield (kg/d) 23.3 21.9 24.0 24.4 23.0 0.37 ** **• NS »

FCM yield (kg/d) 26.5 24.8 27.0 26.4 25.1 0.43 NS •* NS •

Fat yield (g/d) 1146 1069 1159 1109 1064 20 NS ** NS •1146 1069 1159 1109 1064 20 NS •* NS •

Protein yield (g/d) 715 682 757 755 706 13 • •• NS
Milk composition

Fat (g/kg) 49.1 49.0 48.2 45.5 46.4 0.60 *** NS NS NS
Protein (g/kg) 30.6 31.0 31.4 30.8 30.8 0.42 NS NS NS NS

Live weight
Mean (kg) 512 511 518 514 512 2.2 NS NS NS NS
Change (kg/d) 0.16 0.15 —0.19 —0.06 0.20 0.14 NS NS NS NS

Feed conversion
FFU/kg FCM 1 0.383 0.358 0.366 0.359 0.351 0.013 NS NS NS NS
kg DM/kg FCM 0.617 0.616 0.580 0.592 0.625 0.009 NS NS NS **

DCP g/kg FCM 2 60.6 59.9 55.7 56.7 60.7 1.00 NS NS NS *•

Production feed units; intake corrected for maintenance and live weight change
2 Production DCP; intake corrected for maintenance
For significance: see Table 3.

(Table 5). The effect of molasses on produc-
tion was similar, whether included in the Ba
or SBP diet. But cows given SBP supplements
yielded 1.1 kg (P < 0.01) more milk and 32 g
(P < 0.05) more protein on average than those
given Ba supplements. The more complex
mixture of concentrate (treatment C) pro-
duced more milk, FCM, milk fat (P < 0.05)
and milk protein (P < 0.01) than the other
diets. Treatment effects on milk, FCM, fat
and protein production were similar in heifers
and older cows; P for square X treatment in-
teraction was 0.91, 0.33, 0.76 and 0.57, re-
spectively. The protein concentration in the
milk was not affected by the diet, but SBP
supplements decreased fat concentration in
the milk significantly (P < 0.001) compared
to Ba supplements.

In cows given diets C and D the live weight
change was negative, while for those given
other diets it was positive, but the differences
were not statistically significant.

Feed conversion, expressed in terms of pro-
duction feed units per kg FCM, tended to be
slightly better on SBP diets than on Ba. In
terms of DM/kg FCM, there were no differ-
ences between Ba and SBP diets. On supple-

ment C, significantly less (P < 0.01) DM and
DCP were consumed per kg of FCM com-
pared to the other treatments.

The fatty acid composition of milk is shown
in Table 6, with the individual fatty acids
grouped according to their origin. Concentra-
tion of butyric acid (C 4:0) was lower
(P < 0.01) on SBP diets than on Ba diets.
Short-chain fatty acids (C6:O to C,4:1 ) synthe-
sized de novo in the udder were not affected
by the type of supplement. There were no dif-
ferences in the proportions of C,8:0 and C, g:l

which are entirely derived from blood. The
difference between the Ba and SBP diets in
the proportion of linolic acid (C ]B:2), al-
though significant, was small. The proportion
of C-16 acids, which arise partly from blood,
was higher (P < 0.01) on Ba than on SBP
diets. Compared to the Ba diets the SBP diets
increased significantly (P < 0.01) the propor-
tion of fatty acids with odd number of car-
bon. This was mainly due to the difference of
0.30 %-unit (P < 0.001) in the proportion of
C l 5 acid between Ba and SBP diets. Inclusion
of molasses in the diet reduced (P < 0.05) the
proportion of fatty acids with odd number of
carbon.

107



Table 6. Fatty acid composition (°7o by weight) of milk fat of cows given grass silage with different energy supplements.

Statistical significance
of effectTreatmentFatty acid SEM

A B C D E 24 df AB v. AD v. AE v. C. v.
DE BE BD other

4:0 3.65 3.60 3.60 3.34 3.47 0.07 ** NS NS NS
6:0 to 14:1 24.2 24.1 23.8 24.6 23.4 0.73 NS NS NS NS

16:0 to 16:1 35.2 34.9 34.4 32.8 33.2 0.64 *» NS NS NS
18:0 10.2 10.6 10.7 10.7 10.8 0.25 NS NS NS NS
18:1 19.7 20.0 20.3 20.6 22.0 0.96 NS NS NS NS
18:2 2.36 2.32 2.39 2.59 2.46 0.06 ** NS NS NS
18:3 0.58 0.63 0.64 0.65 0.67 0.03 NS NS NS NS

All unsaturated
fatty acids 26.0 26.4 26.4 27.1 28.7 1.10 NS NS NS NS

Fatty acids with
odd number of C 3.54 3.42 3.45 3.98 3.66 0.10 ** * NS NS

For significance: see Table 3.

Table 7. Mean chemical composition (g/kg DM) of experimental diets and feed intake in digestibility trial.

Treatment ...

SEM Sigmti-
A B C D E 12 df cance

Ash 78 93 89 89 99 0.44
Crude protein 161 163 162 163 164 1.09 NS
Crude fibre 214 205 228 259 238 2.54 •**

Ether extract 45 42 44 46 43 0.55 **

NFE 502 495 476 444 453 4.08 ***

NDF 406 378 431 498 445 3.47 •**

ADF 261 250 281 321 294 2.88 ***

ADL 28 26 29 32 29 0.10 ***

Cellulose 234 224 252 290 266 2.88 ***

Hemicellulose 145 128 150 176 151 1.35 ***

DM intake kg/d 17.40 15.88 16.40 16.47 16.35 0.35 NS

For significance: see Table 3.

Digestibility

Table 7 gives the chemical composition of
the experimental diets of the cows during di-
gestibility measurements. All the differences
between the diets were significant except the
concentrationof crude protein. The main dif-
ferences between the Ba and SBP diets were
a higher concentration of soluble residues
(OM crude protein ether extract
NDF) (320 v. 228 g/kg DM) and a lower con-
centration of fibre components in the Ba diets.

There were no significant differences
between the treatments in digestibility of OM
(Table 8). However, there was a trend for

higher OM digestibility with the diets con-
taining molasses. Crude protein digestibility
was significantly (P < 0.001) lower in cows
offered SBP than in those offered Ba supple-
ments. Replacement of Ba by SBP increased
the digestibility of crude fibre (P < 0.01),
NDF, ADF, cellulose and hemicellulose (P <

0.001). Replacing 330 g/kg of barley or SBP
by Mo improved (P < 0.05) the digestibility
of DM and nitrogen-free extracts (NFE). The
digestibility of the various fibre components
was not affected by inclusion of Mo in the
diet. The effect of molasses on the digestibility
of dietary constituents was the same whether
given with barley or SBP. Digestibility of

108



Table 8. Digestibility of different dietary constituents in cows given grass silage with different energy supplements.

Statistical significance
Treatment gEM of effect

A B C D E 24 df AB v. AD v. AE v. C. v.
DE BE BD other

Dry matter 0.720 0.739 0.730 0.707 0.724 0.0062 * * NS NS
Organic matter 0.740 0.755 0.750 0.730 0.741 0.0061 NS NS NS NS
Crude protein 0.728 0.738 0.725 0.691 0.697 0.0085 *** NS NS NS
Ether extract 0.735 0.769 0.760 0.738 0.726 0.0092 NS NS NS NS
Crude fibre 0.623 0.640 0.667 0.684 0.682 0.0121 ** NS NS NS
NFE 0.794 0.805 0.796 0.769 0.788 0.0066 ** * NS NS
NDF 0.600 0.592 0.653 0.682 0.677 0.0119 *** NS NS NS
ADF 0.555 0.551 0.592 0.620 0.622 0.0136 NS NS NS
Cellulose 0.664 0.674 0.705 0.722 0.725 0.0103 *** NS NS NS
Hemicellulose 0.680 0.670 0,762 0.794 0.782 0.0135 *** NS NS NS

For significance: see Table 3.

Table 9. Calculated energy balance (MJ/day) and efficiency of conversion of ME surplus to maintenance into milk
in cows given grass silage and different energy supplements.

Statistical significance
of effectTreatment SEM

A B C D E 24 df AB v. AD v. AE v. C. v.
DE BE BD other

OM intake (kg/d) 15.14 13.93 14.31 14.42 14.27 0.20 NS ** • NS
DOM 1 intake (kg/d) 11.20 10.50 10.72 10.50 10.57 0.15 * * • NS
DE intake 2 210.4 197.4 201.5 197.3 198.7 2.79 * * * NS
ME intake 181.0 169,7 173.3 169.7 170,9 2.34 * � * NS
ME from change of

live weight 3 —5.6 —5.7 +4.3 + 1.0 —7.3 5.23 NS NS NS NS
Energy output

Maintenance 48.9 48.8 49.4 49.0 48.9 0.16 NS NS NS NS
Milk 79.7 74.8 81.8 80.1 76.1 1.18 NS NS

Efficiency
Including live
weight change 0.632 0.660 0.650 0.671 0.667 0.023 NS NS NS NS
Ignoring live
weight change 0.602 0.622 0.668 0.667 0.628 0.013 * NS * *

NE/DE 4 0.546 0.559 0.549 0.567 0.565 0.014 NS NS NS NS

1 DOM = digestible organic matter; 2 DE = digestible energy
' Allowing 28 MJ for each kg lost and subtracting 34 MJ for each kg gained
4 NH = net energy for milk, live weight gain (20 MJ/kg) and maintenance (0.3 MJ/kg W 0 75 )

For significance: see Table 3

silage OM and crude fibre, calculated as a dif-
ference using the digestibility coefficients of
Salo et ai (1982) for concentrates, was not
affected by the treatment. The average diges-
tibilities of silage OM for treatments A E
were 0.671, 0.679, 0.686, 0.669 and 0.671
(SEM 0.011) and those of crude fibre 0.673,
0.681, 0.687, 0.675 and 0.680 (SEM 0.015),
respectively.

Energy balance

Dietary inclusion of 2 kg of molasses
reduced OM intake significantly (P < 0.01)
(Table 9). Digestible organic matter (DOM),
DE and ME intakes were lower (P < 0.05) on
SBP than on Ba diets and lower (P < 0.05)
on diets containing 2 kg of molasses than on
diets without molasses. The effect of molas-

-109



Table 10. Total mean retention time (TMRT), pool mean retention time (PMRT), residual mean retention time
(RMRT), transit time (TT) and rate of passage (k, and k 2) of Cr-labelled straw and CoEDTA in cows
fed on grass silage with different energy supplements.

Treatment SignificanceSEM
A B C D E 12 df Cow Period Treatm.

TMRT 1 Cr (h) 37.7 37.3 37.1 36.8 37.7 0.82 *** NS
PMRT Cr (h) 22.5 21.6 22.1 22.5 22.5 0.85 NS **» NS22.5 21.6 22.1 22.5 22.5 0.85 NS *** NS
RMRT Cr (h) 15.3 15.7 15.0 14.2 15.2 0.89 ** NS NS
TT Cr (h) 9.5 9.6 8.5 9.3 9.4 0.50 * NS NS9.5 9.6 8.5 9.3 9.4 0.50 * NS NS
TMRT 2 Cr (h) 37.5 37.3 36.8 37.2 37.9 0.80 •** *** NS
k, Cr 0.047 0.048 0.047 0.045 0.046 0.002 NS *** NS0.047 0.048 0.047 0.045 0.046 0.002 NS *** NS
k 2 Cr 0.215 0.188 0.188 0.215 0.196 0.021 ** ••• NS
TMRT Co (h) 19.9 19.6 18.3 18.3 18.0 0.66 *» * NS
PMRT Co (h) 11.0 9.7 9.8 9.7 10.3 0.47 * *** NS
RMRT Co (h) 9.0 9.8 8.5 8.6 9.2 0.43 ** �** NS
k, Co 0.096 0.108 0.104 0.105 0.099 0.004 * *** NS

Calculated according to TiELEMANset at. (1978);
2 Calculated according to Grovum and Phillips (1973); For significance see Table 3.

ses on estimated DOM intake was different
(P < 0.05) when given with barley than when
given with SBP; there was a decrease of 0.7
kg when given with barley, whereas no effect
was found withSBP. Production of milk ener-
gy was lower (P < 0.001) with inclusion of
2 kg of molasses. The efficiency of transfer-
ring surplus ME into milk (kl0) averaged 0.656
when live weight change was included, and
was not affected by the energy source. Ignoring
live weight change, k, was higher (P<0.05)
on the SBP diets than on the Ba diets, and
higher for diet C than the other diets. When
included in barley, molasses improved k„
while on SBP diet the reverse was true. The
interaction was significant (P < 0.05) when
live weight change was ignored.

Digesta passage

No significant treatment effect was found
on the measurements of retention time or pas-
sage rate of particle or liquid phase of the
digesta (Table 10). On the other hand, most
of the cow and period effects on digesta pas-
sage parameters were significant. Estimates
for TMRT of Cr-straw were similar, irrespec-
tive of the calculation method: the value for
TMRT was 37.3 (SE 0.35) when calculated ac-

cording to Grovum and Phillips (1973) and
37.3 (SE 0.35) according to Thielemans el
al. (1978). The relation between these two
methods was:

y = 1.2 (SE 1.8) + 1.03 (SE 0.047)
X (RSD 1.14; r 0.977)

Higher DM intake as per cent of live weight
(Table 11) decreased the TMRT of particle
marker (P < 0.001) and liquid marker
(P < 0.05). DM intake did not affect the pool
mean retention time (PMRT) significantly,
but at higher levels of intake the residual mean
retention times (RMRT) of particles and liquid
were decreased (P < 0.001; P < 0.01).

Table 11. The regression on digesta retention (h) on DM
intake (°/o of live weight).

Param- Intercept Slope Signifi- RSD
eter (SE) (SE) cance

TMRTCr 58.9 (4.5) —7.00 (1.43) *** 2.38
PMRT Cr 24.2 (3.2) —0.63 (1.04) NS 1.71
RMRTc, 34.8 (3.5) —6.37 (1.13) 1.86
TMRTCo 28.8 (4.2) —3.14 (1.29) * 2.15
PMRT Co 11.9 (2.5) —0.59 (0.80) NS 1.32
RMRT ( „

16.9 (2.2) —2.56 (0.71) •• 1.07

SE = standard error;
RSD = residual standard deviation

110



Discussion

The silage given to the cows was well pre-
served with low pH and low concentration of
fermentation acids, with 0.76 of the fermen-
tation acids being lactic acid. The extent of
proteolysis was small with low proportions of
ammonia and soluble N in total N.

Feed intake

There was no difference in silage DM in-
take between the Ba and SBP supplements, in
agreement with the results of Castle et al.
(1981). They reported decreases in the intake
of silage DM of 0.44and 0.40 kg respectively,
when feeding 1 kg of Ba or SBP. On the other
hand, Thomas et al. (1986) recorded a signif-
icant increase of 0.9 kg in silage DM intake
when Ba supplement was replaced by a mix-
ture of unmolassed beet pulp (555 g/kg) and
rice bran (314 g/kg). In view of the similar
substitution rate they suggested that the in-
crease in forage intake was not mediated solely
by changes in fibre digestion in the rumen.
Chamberlain et al. (1984) found a different
response in silage intake when they modifyed
the composition of supplement, and their data
indicate that higher intakes with SBP can be
expected only with highly digestible silage and
high concentration of total N. In contrast to
this, Mayne and Gordon (1984) reported
with highly digestible silage (D-value 0.691) a
significantly lower silage DM intake on SBP
than on Ba. In the present study the D-value
of silage was 0.622, similarly to the studies of
Castle et al. (1981) and Thomas et al.
(1986). ISTASSE et al. (1986) found a tendency
for lower feed intake on SBP than on Ba
supplements on diets based on NH,-treated
straw.

Under other circumstances, in grazing cattle
(Meus 1986) or with high levels of concen-
trate in hay based diets (Sutton et al. 1984),
forage intakes has been higher in cows given
concentrates based on fibrous by-products
than in those given starchy concentrates. A
possible explanation of the increased forage
intake is given by Steg et al. (1985). High

4

levels of easily fermentable carbohydrates in
starchy concentrates tend to decrease rumen
pH and increase VFA concentration in the
rumen, resulting in a lower cellulolytic activi-
ty. However, in the present study, the supple-
ments comprised ca. 0.43 of total DM, and
on that level of supplementation to cows given
silage based diets Murphy et at. (1986) ob-
served no differences in rumen pH between
barley and barley-sugar beet pulp (1:2) sup-
plements. Similarly, there were no differen-
ces in the rate of disappearance of silage or
hay DM from nylon bags, and rumen pH did
not differ in bulls given diets of silage supple-
mented withbarley or SBP at the level of 0.52
of the total DM intake (Huhtanen 1987).

The effect of dietary inclusion of molasses
on silage DM intake cannot be explained
solely by physical regulation. Voluntary intake
of rations rich in forages is generally restrict-
ed by the capacity of thereticulorumen. Dis-
appearance of digesta from the rumen occurs
either by microbial degradation and absorp-
tion of end products or by passage of un-
digested residues to the lower digestive tract
(Steg et at. 1985). However, in the present
study, molasses had no effect either on the
digestibility of different cell wall components
or on the retention time of Cr-labelled straw
particles. The degradation rate of forages in
the rumen was similar in bulls when 330 g/kg
of Ba was replaced by Mo (Huhtanen 1987).
Possibly the lower forage intake on Mo diets
is due to metabolic control. In the case of
molasses containing diet, the higher fermenta-
tion rate of sugars than that of starch or fibre
(Sutton 1980) may increase the concentra-
tions of ruminal acetate, duodenal lactate and
hepatic uptake of propionate, which are feed-
back signals for termination of feeding
(Forbes 1980). Because the access to feed was
limited to 3 h per feeding time, early termi-
nation of the meal may have reduced forage
intake in cows given Mo diets.

Production response
Cows given SBP yielded more milk than

those given Ba either with or without molas-

111



ses. A similar increase in milk yield was also
recorded by Thomas et al. (1986) when Ba
supplement was replaced by a mixture of SBP,
extracted rice bran and fat prills. However,
Mayne and Gordon (1984) found no change
in milk yield with Ba and SBP used at two
levels of supplementation with ad libitum
silage and Istasse et al. (1986) found a ten-
dency for higher milk yield on Ba based diet
than on SBP based diet. In earlier experiments
(Bhattacharya and Sleiman 1971, Bhat-
tachryaand Lubbadah 1971,Castle 1972,
Castle et al. 1981) there were no statistical-
ly significant differences in milk production
between Ba and SBP supplements, although
in most studies the milk yield was slightly
higher when SBP diets were given. In grazing
cows the milk yield was higher when Ba was
replaced by SBP (Murphy 1985), or more
generally, starchy concentrate by fibrous con-
centrate (Meus 1986). The experimental de-
sign of the present study does not allow as-
sesment of the residual effects of the diets, but
Thomas et al. (1986) found in mid lactation
a large positive residual effect of + 1.6kg per
day of substitution of Ba with SBP based
supplement.

Milk fat content was lower in cows given
SBP supplements than in those given Ba. The
linear trend of the level of SBP (kg DM/d)
explained 94 % of the sum of squares of the
treatment effect. This is in agreement with the
observation of Thomas et al. (1986) of a
markedly higher fat content in cows given Ba
supplement at a level of supplement similar
to the present study (ca. 0.44). Similarly,
Castle and Watson (1975) found a lower
milk fat content when barley was replaced by
dried grass as a supplement to silage, despite
a higher forage intake and fibre content in the
dried grass diet. On the other hand, at a level
of ca. 0.4 of the total DM intake, Castle et
al. (1981) noticed no differences in milk fat
content between Ba and SBP supplements.

Only small differences were found at the
higher level of supplement (0.65) in milk fat
content (Thomas et al. 1986). Likewise,
Mayne and Gordon (1984) and Sutton et

al. (1985) observed little effect of the type of
supplement on milk fat when the supplement
comprised 0.5 to 0.6 of the total DM. How-
ever, marked reductions in milk fat content
and fat yield were noted when starch based
concentrate was fed at the level of 0.8 of total
DM (Sutton et al. 1985) or SBP replaced by
flaked maize (Lees et al. 1982).

In the present study, the reduced milk fat
content on SBP diets cannot be explained by
theproportion of forage in the diet or by die-
tary fibre content, which exceeded the mini-
mum recommended value of 18 % of crude
fibre (Kaufmann 1976) or 22 % of ADF
(Sutton 1984), and was higher on SBP diets.

According to SUTTON (1984), changes in
the rumen VFA proportions account for
about two thirds of the variation in milk fat
content and, on average, there is a fall in milk
fat content of approximately 5 g/kg for every
unit fall in the ratio acetate plus butyrate to
propionate. Rumen VFA was not measured
in the present study, but Murphy et al. (1986)
found a decrease in this ratio from 5.2 to 4.5
in dairy cows when 670 g/kg of Ba was re-
placed by SBP. The results of Castle (1972)
indicate a similar trend, and Castle noted as
well a lower fat content on SBP diet than on
Ba diet. In these studies, the major changes
in rumen VFA proportions included an in-
crease in propionate and a decrease in butyr-
ate on SBP diets. When different VFAs were
infused into the rumen, propionate had a
negative effect and butyrate a highly positive
effect on milk fat content (Thomas and
Chamberlain 1984). In the present study
VFA theory finds support in the trend toward
a lower concentration of /3-hydroxybutyrate
in blood plasma (Miettinen and Huhtanen
1987) and in the lower proportion of butyric
acid in milk fat in cows given SBP supple-
ments. In grazing cows, on the other hand,
the proportion of butyrate tended to be higher
and that of propionate lower when barley was
replaced by molassed SBP (Murphy 1985).

In silage diets, inclusion of cereals has been
found to reduce the proportion of acetate and
increase the proportion of butyrate, and pro-

112



duce only small changes in the proportion of
propionate (Thomas et al. 1980). This may
explain the high concentration of milk fat
when barley diets were given.

A significantly higher protein yield on the
SBP diet than on the Ba diet has not been
reported earlier, but a similar trend has been
observed by Castle et al. (1981), Thomaset
al. (1986) and ISTASSE et at. (1986). In the
present study, the increase in protein yield
with SBP supplements was accompanied by
an increase in milk yield without any change
in protein contents. The similar protein content
with barley and SBP diets is in agreement with
the findings of CASTLE (1972), CASTLE et al.
(1981) and Mayneand Gordon (1984). The
findings of Thomas et al. (1986) of a lower
concentration of milk protein with SBP sup-
plement may have been due to the inclusion of
fat in their SBP diet. Two suggestions can be
offered for higher protein yield with SBP.
First, the proportion of protein of SBP de-
graded in the rumen has been reported to be
lower than that of barley (Mayne and Gor-
don 1984, Madsen and Hvelplund 1986).
The effect of increased flow of dietary pro-
tein depends, however, on the balance
between the protein supply and requirement.
Second, of the individual end products of
digestion, acetate and propionate increased
and butyrate decreased the milk protein yield
in the infusion studies reviewed by (Thomas
and Chamberlain 1984). Thus, a high pro-
portion of butyrate in rumen VFA on silage
diets supplemented with cereals (Thomas et.
al. 1980) may explain the lower protein yield
on Ba diets.

Inclusion of 2 kg of molasses in Ba and SBP
caused a significant and similar reduction in
milk yield (1.4 kg) without any change in milk
composition. In the case of Ba diet, the de-
creased production can probably be attributed
to the lower silage intake with Mo. The reduc-
tion in milk yield on SBP diet is in agreement
with recent observations of Parkins et al.
(1986) and Hemingway et al. (1986) who
found that the inclusion of molasses in SBP
in silage based diets generally enhances the

content of milk fat, but reduced milk yield.
In the present study, the slightly lower silage
intake is insufficient to explain the reduced
milk yield in cows given SBP and molasses.
Rather the tendency for higher live weight gain
when molasses was included in SBP suggests
a change in nutrient partition towards body
tissues.

Lower efficiency of utilization of large
amounts of soluble sugars has been reported
by Broster et al. (1970) who compared 2.7
kg of sucrose and 3.2 kg of glucose with equal
amounts of digestible energy of concentrate,
by Lofgreen and Otagaki (1960) who ad-
ded 2 kg of molasses and by Bowman and
Huber (1967) who used 3.5 kg of lactose. A
comparison of fodder beet with Ba showed
beet to maintain higher fat content than Ba
but to result in lower yields, so that fat yield
was the same (Krohn and Andersen 1980).

The increase in milk, FCM, milk fat and
protein yields in response to feeding the more
complex supplement (treatment C) can part-
ly be explained by the negative effect of high
levels of Mo in the diet. However, supplement
C also produced a small positive effect of
+ 0.5 kg of FCM, + 31 g of milk fat and +

22 g of milk protein per day compared to the
mean values for Ba and SBP given without
Mo. The live weight loss in cows given sup-
plement C tended to be greater, suggesting an
increase in the amount of energy withdrawn
from body tissues. The slightly higher con-
centration of acetoacetate in cows given sup-
plement C (Miettinen and Huhtanen 1987)
also suggests increased mobilization of body
fat. Energy balance experiments have shown
that the productive value of a mixed diet is
in many instances greater than the arithmetic
mean of the ingredient feedstuffs (MOEet al.
1972). Although the full explanation of this
associative effect is not clearly understood, it
is likely that at least a portion of the effect
is due to nutritional imbalanceof one or both
ration components when fed alone. In the pre-
sent study, the associative effect of Ba and
SBP and the effect of a lower level of Mo are
confounded, but it is unlikely that 1 kg of mo-

113



lasses would have a positive and 2 kg of mo-
lasses a great negative effect on the milk yield.

Only minor, though sometimes significant,
differences in milk fatty acid composition
were found between Ba and SBP supplements.
Yields of individual fatty acids reflected the
total fat yield. The lower proportion and yield
of butyric acid on SBP than on Ba diets were
related to lower concentration of plasma /3-hy-
droxybutyrate (1.76 v. 2.30 mM). According
to the review by Mooreand Christie (1981),
about 40 % of the butyric acid in milk is syn-
thesized by condensation of two acetate mo-
lecules, while the remaining 60 % appears to
be derived from C 4 components.

The Ba and SBP supplements had a similar
effect on the proportion and total yield of
C6—C l 4 fatty acids, suggesting that the avail-
ability of precursors, i.e. VFA, in the rumen
was not affected by the type of supplement.
An increase in the proportion of fatty acids
of odd carbon number, especially of Cls, on
the SBP diet mayreflect an increased synthesis
of microbial lipids in the rumen. Mixed rumen
bacteria and protozoa contain a high propor-
tion of fatty acids of odd carbon number,
both straight and branched chain (Harfoot
1981). The higher proportion of Cl 5 with
SBP cannot be explained by dietary origin, be-
cause unmolassed SBP contains only 0.13 —

0.18 % of digestible ether extract (Kelly
1983). Another primer to give with fatty acids
with odd number of carbon atoms is propio-
nate, which may be incompletely metabolized
in the liver and enters the peripheral circula-
tion with diets asssociated with high propionate
production (Rook and Thomas 1983).

Digestibility

The similar OM digestibilities of the Ba and
SBP diets agree with the results of Mayne

and Gordon(1984). However, Thomas et al.
(1986) reported a lower digestibilities of DM
and gross energy for SBP diet. The lower
crude protein digestibility with SBP may be
due to the heating during the drying process,

and in part to the higher proportion of ADF
bound nitrogen in SBP than in cereal grains.

The lower digestibility of fibre components
on barley diets is in agreement with many
other studies comparing starch and fibre based
supplements, and has been attributed to
reduced cellulolysis in the rumen with high
levels of starch (El-Shazly et al. 1961). The
level of supplementation has been found to
influence the effect of type of concentrate on
fibre digestion (Thomas et al. 1986): when
the level barley was increased the digestibility
of fibre decreased markedly, but very little
when the level of SBP was increased.

The high digestibility of SBP fibre (Bhat-
tacharya and Sleiman 1971) contributed all
the difference in the digestibility of fibre com-
ponents, and the digestibility of silage was
unaffected by the type of supplement. This
result is consistent with the findings in the
nylon bag incubations in bulls given similar
diets (Huhtanen 1987). In contrast, Istas-
se et al. (1986) reported that a concentrate
based on SBP did not affect the digestibility
of NH 3 -treated straw, whereas Ba depressed
the digestibility of straw from 0.559 to 0.393
in dairy cows. A similar difference in effect
was observed with diets for sheep of 0.3 of
NH 3 -treated straw and 0.7 of rolled Ba or
SBP (Fahmy et al. 1984). The lower propor-
tion of supplement (0.43) in the present study
may be one reason for this discrepancy. Al-
so, the type of forage may be relevant to the
extent of reduction in fibre digestion in the
rumen when starch or fibre based supplements
are fed. Thus, an increase in the proportion
of Ba based concentrate from 0 to 0.6 in the
diet of dairy cows reduced the degradability
of NHj-treated straw more than that of grass
silage (Fahmy and Sunstol 1985).

Inclusion of Mo in the Ba diet did not im-
pair the digestibility of fibre or silage OM.
Supplementation of silage diets with sucrose,
however, depressed the cellulose digestibility
in calves (England and Gill 1985). Similar-
ly, Mertens and Loften (1980) found that
the addition of starch to forage based diets
leads to a reduction in cellulolytic activity as

114



a result of rapid starch fermentation in the
rumen. The results of the present study in-
dicate that sugars do not depress fibre diges-
tion when they are used to replace starch
rather than as a supplement in silage based
diet. The same conclusion can be drawn from
the replacement of SBP by Mo. At higher le-
vels of supplementation or inclusion of
molasses, however, the response may be dif-
ferent.

Utilization of energy

Calculations of energy utilization confirm
the equal or slightly better productive value
of SBP than of barley. Ignoring live weight
change the efficiency was higher (P<0.05)
with SBP diets, but including live weight
change the differencedid not reach statistical
significance, mainly because of the great stand-
ard error in live weight change. According to
Alderman et al. (1982) live weight change is
a poor method of estimating changes in body
energy; for cows in early lactation at zero
energy balance they found a live weight chan-
ge of +0.42 kg. According to Thomas and
Rook (1983), most changes in diets which en-
hance milk output without a concomitant in-
crease in energy intake reflect a repartitioning
of energy. However, analysis of blood meta-
bolites (Miettinen and Huhtanen 1987) has
not indicated any great difference in the par-
titioning of energy between the SBP and Ba
diets. The difference in ME output and ME
supply between SBP and Ba diets was +5.5
MJ per day greater on the SBP than on Ba
diets, which is equivalent to 1.1 kg of FCM
or 0.16 kg of live weight gain. Relative to the
standard deviation of live weight change (0.44
kg/d) and of the difference between the ob-
served ME supply and calculated ME output
(14.6 MJ), this difference is small.

Certain assumptions were made in the cal-
culation of energy utilization. A lactose con-
tent of 48 g/kg milk was assumed for all diets
in calculating the milk energy, in agreement
with other studies where SBP and Ba have
been compared. Likewise, a similar energy

value of DOM and factor to convert DE to
ME were used for all diets. Van Es (1971) ob-
served a slightly higher energy valuefor DOM
of barley than for DOM of SBP, probably due
to the higher fat and lower sucrose content of
barley. However, in the present study the
lower fat content of SBP was balanced with
rapeseed oil. No differences in the ratio
DE/ME were found between barley and SBP
or when starch was replaced by sucrose in diets
given to dairy cows (Van Es et al. 1971).

Similar small increases in the efficiency of
use of ME for milk production were noted by
Van der Honig and Wieman (1983) when
starchy and fibrous concentrates were com-
pared in a calorimetric study. Van Es et al.
(1971) reported that DE of Ba and SBP was
used with the same efficiency. Tyrrel et al.
(1973) found that a greater proportion of the
increase in energy balance was in milk when
SBP was added to diet than when corn was
added. Similar changes in energy partition oc-
cur when the ratio forage to concentrate is in-
creased.

From the calculations of energy utilization,
it can be concluded that the lower milk yield
with diets containing 2 kg of molasses was not
entirely caused by a change in the efficiency
of the use of ME. On Ba diet forage intake
was decreased, and on SBP diet live weight
gain tended to be increased. In an energy
balance trial Van Es et al. (1971) did not
notice differences in the utilization of DE
when 8 % sucrose in concentrate was replaced
by 8 % starch. In the present study, sugar con-
tent was higher, 108 and 118 g/kg DM, for
treatments B and E, respectively.

Digesta passage

The model used to describe marker excre-
tion accounted for an average 93 % of the
variance in faecal marker concentration,
which is slightly less than found with a more
sophisticated model by Dhanoa et al. (1985).
The graphical procedure used here gave a poor
fit for early values of marker excretion curves.
However, there were no differences in TMRT

115



when calculated as TMRT = TT + \/k { + 1/k2
or from the relative marker concentrations-
sampling interval according to Thielemans
et al. (1978). Similar values for TMRT have
been reported with the grab sampling and total
faeces collection methods, for both dairy cows
(Hartnell and Satter 1979) and sheep
(Lindberg 1985).

The values reported here for TMRT,
PMRT of liquid phase and the intercept at
zero intake are similar to those reported by
Uden (1984). Hartnell and Satter (1979)
report a value of 26—28 h for liquid TMRT.
The present value of 37 h for TMRT for solids
is low compared to the values of about 50 h
reported by Hartnell and Satter (1979) and
Uden (1984). In the present study, the re-
sidual mean retention time was considerably
shorter than that reported by Uden (1984),
but the pool mean retention time was only
2—3 h shorter. The discrepancy may be at-
tributed to the different physical character-
istics of theparticulate marker. In the present
study, the Cr-straw was ground to pass a
1,5-mm screen so that particles were available
for passage without further comminution. The
effect of the production of small particles
from the large particle pool due to chewing,
rumination and microbial activity was not in-
cluded in TMRT, and hence only reflects
the turnover of the small particle pool. On the
other hand, anything which alters the char-
acter of fibre or inhibits microbial activity
would affect the rate of production of small
particles and confoundany analysis of the rate
of passage from the reticulorumen (Grovum
1984). In this regard Cr-mordanted food
would appear to be unacceptable. The sig-
nificant period effect on particulate passage

may result in part from the differences in
volume weight of Cr-straw between periods
(Ehle et al. 1984).

Conclusions

In conclusion, the results show that SBP
given on equal DM basis results in higher milk
and protein yields than does barley. Because
of the lower fat content on the SBP diet, FCM
and milk fat yields were not affected. Ac-
cording to this study, the different sources of
digestible energy were utilized with similar ef-
ficiency. Therefore, instead of regarding 1 kg
of DM of SBP as equivalent to 0.87—0.90 kg
of DM of barley in the Finnish Feed Tables
(Salo et ai. 1982), SBP should replace barley
on equal DM basis in the feeding of dairy
cows. At the level of supplementation used in
the present study (0.43 of total DM), no ef-
fect can be expected on the silage DM intake
or on the digestibility of silage when barley
is replaced by SBP. Including high levels of
molasses in Ba or SBP leads to a reduction
in milk yield without any marked change in
milk composition. The best production re-
sponse was obtained with a low level of mo-
lasses given with a mixture of barley and SBP
indicating a synergism from the fermentation
of ingredients at different rates.

Acknowledgement. The author isgrateful to Mrs. Han-
nele Ala-Seppälä, Mr. Kari Tahvanainen and Mr. Mik-
ko Ranta for excellent technical assistance during the ex-
periments, to Mr. Martti Himanen and his staff for taking
care of the animals, and to Finn Sugar Ltd. for financial
support and for providing sugar beet pulp and molasses.
The analyses of milk fat and protein content by Valio
Laboratories and milk fatty acids by the State Institute
of Dairy Research are greatfully acknowledged.

References

Alderman, G., Broster, W.H., Strictland, M.J. &

Johnson, C.L. 1982. The estimation of the energy
value of liveweight change in lactating dairy cows.
Livest. Prod. Sci. 9: 665—673.

Arc 1980. The Nutrient Requirements of Ruminants.
Commonwealth Agricultural Bureaux. 351 p. London.

Bhattacharva, A.N. & Sleiman, F.T. 1971. Beet pulp
as a grain replacement for dairy cows and sheep. J.

116



Dairy Sci. 54: 89—94.
& Lubbadah, W.F. 1971. Feeding high levels of beet

pulp in high concentrate dairy rations. J Dairy Sci. 54:
95—99.

Barker, S.B. & Summerson, W.H. 1941. The colori-
metric determination of lactic acid in biological ma-
terial. J. Biol. Chem. 138: 535—554.

Bowman, R.L. & Huber, J.T. 1967. Effect of dietary
lactose on milk composition and rumen volatile fatty
acids. J. Dairy Sci. 50: 579—581.

Broster, W.H., Sutton, S.D., Smith, T., Broster, V.J.
& Balch, C.C. 1970. The effect of supplements of
sucrose and of glucose monohydrate on the milk pro-
duction and live weight of dairy cows. J. agric. Sci.,
Camb. 74: 217—225.

Chamberlain, D.G., Thomas, P.C., Wilson, W.D., Kas-
sem, M.E. & Robertson, S. 1984. The influence of the
type of carbohydrate in the supplementary concentrate
on the utilization of silage diets for milk production.
Proc. 7th Silage Conf. Queen’s Univ., Belfast. (Ed.
Gordon, F.J. & Unsworth, E.F.). pp. 37—38.

Castle, M.E. 1972. A comparative study of the feeding
value of dried sugar-beet pulp for milk production. J.
agric. Sci., Camb. 78: 371 —377.

—, Gill, M.S. & Watson, J.N. 1981. Silage and milk
production: a comparison between barley and dried
sugar beet pulpas silage supplement. Grass Forage Sci.
36: 319—324.
- & Watson, J.N. 1975. Silage and milk production:
a comparison between barley and dried grass as sup-
plements to silage high digestibility. J. Br. Grassld. Soc.
30: 217—222.

Dhanoa, M.S., Siddons, R.C., France, J. & Gale, D.L.
1985. A multicompartmental model to describe marker
excretion patterns in ruminant faeces. Br. J. Nutr. 53:
663—671.

Ehle, F.R., Bas, F., Barno, 8., Martin, R. & Leone,
F. 1984. Particulate rumen turnover rate measurement
as influenced by density of passage marker. J. Dairy
Sci. 67: 2910—2913.

El-Shazly, K., Dehority, B.A. & Johnson, R.R. 1961.
Effect of starch on the digestion of cellulose in vitro
and in vivo by rumen microorganisms. J. Anim. Sci.
20: 268—273.

England, P. & Gill, M. 1985. The effect of fish meal
and sucrose supplementation on the voluntary intake
of grass silage and live-weight gain of young cattle.
Anim. Prod. 40: 259—265.

Es, A.J.H. van, Nijkamp, H.J. & Vogt, J.E. 1971. The
net energy content of dried sugar-beetpulp and of suc-
rose when fed to lactating cows. Neth J. agric. Sci. 19:
48—56.

Fahmv, S.T.M., & Lee, N.H. & orskov, E.R. 1984.
Digestion and utilization of straw. 2. Effect of different
supplements on digestion of ammonia-treated straw.
Anim. Prod. 38: 75—81.
& Sundstol, F. 1985. The degradability of untreated

and chemically treated barley straw and of grass silage
as influenced by the ration composition. Z. Tierphysiol.,
Tierernähr. u. Futtermittelkde. 53: 34—42.

Flatt, W.P., Moe, P.W., Moore, L.A., Hooven, N.W.,
Lehmann, R.P., orskov, E.R. & Hemken, R.W. 1969.
Energy utilization by high producing dairy cow. Ener-
gy Metabolism of Farm Animals (Ed. Blaxter, K.L.,
Kielanowski, J. & Thorbek, G.) EAAP Pubi. No. 12.
pp. 221—251.

Forbes, J.M. 1980. Hormones and metabolites in the
control of feed intake. In Digestive Physiology and
Metabolism in Ruminants. (Ed. Ruckebush, Y & Thi-
vend, P.) pp. 145—159. MTP Press, Lancaster.

Goering, H.R. & Van Soest, P.J. 1970. Forage fiber
analyses. Agriculture Handbook No. 379. United States
Dept, of Agric. Washington.

Grovum, W.L. 1984. Integration of digestion and digesta
kinetics with control of feed intake a physiological
framework for model of rumen function. Herbivore
Nutrition in Subtropics and Tropics. (Ed. Gilchirst,
F.M.C. & Mackie, R. 1.) pp. 244—268. The Science
Press. Graighall.

& Philips, D.G. 1973. Rate of passage studies in
sheep. 5. Theoretical considerations based on a physical
model and computer simulation. Br. J. Nutr. 30:
377—390.
& Williams, V.J. 1973. Rate of passage in sheep. 4.

Passage of marker through alimentary tract and the
biological relevances of rate-constants derived from
changes in concentration of marker in faeces. Br. J.
Nutr. 30; 313—329.

Harfoot, C.G. 1981. Lipid metabolism in the rumen.
Lipid Metabolism in Ruminant Animals. (Ed. Christie,
W.W.) pp. 21—55. Pergamon Press.

Hartnell, G.F. & Satter, L.D. 1979. Determination of
rumen fill, retention time and ruminal turnover rates
of ingesta at different stages of lactation of dairy cows.
J. Anim. Sci. 48: 381—392.

Hemingway, R.G., Parkins, J.J. & Fraser, J. 1986.
Sugar beet pulp products for dairy cows. Anim. Feed
Sci. Technol. 15: 123—127.

Honing, Y. Van der & Wieman, B.J. IVVO-Report 145.
Institute for Livestock Feeding and Nutrition Research:
Lelystad. (ref. Steg, A, Honing, Y. Van der & Visser,
H De).

Huhtanen, P. 1987. The effects of barley, unmolassed
sugar beet pulp and molasses supplements on organic
matter, nitrogen and fibre digestion in the rumen of
cattle given silage diet. Submitted.

Huida, L. 1973. Quantitative determination of volatile
fatty acids by gasliquid chromatography. J. Scient.
Agric. Soc. Finl. 45: 483—488.

Istasse, L., Reid, G.W., Tait, C.A.G. & orskov, E.R.
1986. Concentrates for dairy cattle: effects on feeding
method, proportion in diet and type. Anim. Feed Sci.
& Technol. 15; 167—182.

Kaufmann, W. 1976. Influence of the composition of the

117



ration and the feeding frequency on pH-regulation in
the rumen and on the feed intake in ruminants. Livest.
Prod. Sci. 3: 103—114.

Lees, J.A., Garnsworthy, P.C. & Oldham, J.D. 1982:
The response of dairy cows in early lactation to sup-
plements of protein given with rations designed to pro-
mote different patterns of rumen fermentation. Forage
Protein in Animal Production. Occasional Pubi. No.
6. Brit. Soc. Anim. Prod. (Ed. Thomson, D.J., Beever,
D.E. & Gunn, R.G.) pp. 157—159.

Kelly, P. 1983. Sugar beet pulp a review. Anim. Feed
Sci. & Technol. 8: I—lB.

Krohn, C.C. & Andersen, P.E. 1979. Ratios with beet
or barley fed separately or in complete ration for dairy
cows. Beretnig fra Statens Husdyrbrugsforsog. 480.

Lindberg, J.E. 1985. Retention of chromium-labelled
feed particles and of water in gut of sheep given hay
and concentrate at maintenance. Br. J. Nutr. 53: 559—

567.
Lofgreen, G.P. & Otagaki, K.K. 1960. The net energy

of blackstrap molasses for lactating dairy cows. J. Dairy
Sci. 43: 220—230.

Madsen, J. & Hvelplund, T. 1985. Protein degradation
in the rumen. Acta. Agric. Scand. Suppl. 25: 103—124.

Male (Ministry of Agriculture, Fisheries and Food).
1975. Energy allowances and feeding systems for

ruminants. Technical Bulletin 33. 79 p. London.
Mayne, C.S. & Gordon, F.J. 1984. The effect of type of

concentrate and level of concentrate feeding on milk
production. Anim. Prod. 39: 65—76.

McCullough, H. 1967. The determination of ammonia
in whole blood by direct colorimetric method. Clin.
Chem. Acta 17: 297—309.

Mertens, D.R. & Loften, J.R. 1980. The effect of starch
on forage digestionkinetics in vitro. J. Dairy Sci. 63:
1437—1446.

Meijs, J.A.C. 1986. Concentrate supplementation of
grazing dairy cows. 2. Effect ofconcentrate composi-
tion on herbage intake and milk production. Grass
Forage Sci. 41: 229—235.

Miettinen, H. & Huhtanen, P. 1987. The effect of the
type of supplement on blood metabolites in dairy cows
given grass silage ad libitum. Unpublished.

Moe, P.W., Flatt, W.P. & Tyrrel, H.F. 1972. Net
energy values of feed for lactation. J. Dairy Sci. 55;
945—958.

Moore, J.H. Christie, W.W. 1981. Lipid metabolism in
the mammary gland of ruminant animals. Lipid
Metabolism in Ruminant Animals. (Ed. Christie,
W.W.) pp. 227—277. Pergamon Press, Lancaster.

Murphy, J.J. 1985. Effect of feeding sodium bicar-
bonate in the concentrate or beet pulp on milk yield
and composition in cows after turnout to pasture in
spring. Ir. J. agric. Res. 24: 143—149.

Murphy, M., Huhtanen, P. & Ala-SeppAla, H. 1986.
The effect of different supplements on rumen and total
tract digestion in cows given grass silage. Unpublished.

Nelson, N. 1944. A photometric adaptation of Somogyi
method for the determination of glucose. J. Biol. Chem.
153: 375—380.

Parkins, J.J., Hemingway, R.G. & Fraser, J. 1986. A
note on dried molassed sugar beet pulp and unmolassed
pressed sugar beet pulp as comparative foods for dairy
cows. Anim. Prod. 43: 351—354.

Porter, M.G., Patterson, D.C., Steen, R.W. & Gor-
don, F.J. 1984. Determination of dry matter and gross
energy ofgrass silage. Proc. 7th Silage Conf. Queen’s
Univ. Belfast. (Ed. Gordon, F.J. & Unsworth, E.F.).
pp. 89—90.

Robertson, J.B. & Van Soest, P.J. 1977. Dietary fiber
estimation in concentrate feedstuffs. 69th meeting of
Amer. Soc. of Anim. Sci. 639.

Rook, J.A.F. & Thomas, P.C. 1983. Milk secretion and
its nutritional regulation. Nutritional Physiology of
Farm Animals. (Ed. Thomas, P.C. & Rook, J.A.F.)
pp. 315—368.

Salo, M-L. & Salmi, M. 1968. The determination of
starch by amyloglucosidase method. J. Scient. Agric!
Soc. Finl. 40: 38—45.

—, Tuori, M. & Kiiskinen, T. 1982. Rehutaulukot ja
ruokintanormit. 70 p. Helsinki.

Stec, A., Honig, Y. Van der & Visser, H. De. 1985. Ef-
fect of fibre in compound feed on the performance of
ruminants. Recent Advances in Animal Nutrition. (Ed.
Haresign, W. & Cole, D.J.A.) pp. 113—129. Bul-
terworths.

Sutton, J.D. 1980. Influence on nutritional factors on
the yield and content of milk fat: dietary components
other than fat. Bull. Int. Dairy Fed. 125: 126—134.

1984. Feeding and milk fat production. Milk Com-
positional Quality and its Importance in Future
Markets. Occasional Pubi. No. 9. Br. Soc. Anim. Prod
(Ed. Castle, M. & Gunn, R.G.) pp. 43—52.

—, Bines, J.A., Napper, D.J., Wilkes, J.M. & Schul-
ler, E. 1984. Form of carbohydrates in dairy concen-
trates. Report of National Institute for Research in Dai-
rying. p. 74,

—, Bines, S.A. & Napper, D.J. 1985. Comparison of
starchy and fibrous concentrates for lactating dairy
cows. Anim. Prod. 40: 533. (Abstr).

Thielemans, M.R., Fracois, E., Bodart, C. & Thewis,
A. 1978. Measurement of gastrointestinal transit in pigs
by radiolanthanide. Comparison with sheep. Ann. Biol.
Anim. Biochem. & Biophys. 18: 237—247.

Thomas, C., Aston, K., Daley, S.R. & Bass, J. 1986.
Milk production from silage. 4. The effect of the com-
position of the supplement. Anim. Prod. 42: 315—325.

Thomas, P.C. & Chamberlain, D.G. 1984. Manipulation
of milk composition to meet market needs. Recent Ad-
vances in Animal Nutrition 1984. (Ed. Haresign, W.
& Cole, D.J.A.). pp. 219—243. Butterworths.
& Rook, J.A.F. 1983. Milk production. Nutritional

Physiology of Farm Animals. (Ed. Thomas, P.C. &

Rook, J.A.F.) pp. 558—622. Longman.

118



—, Kelly, N.C., Chamberlain, D.G. & Chalmers, J.S.
1980. Some aspects of energy and protein utilization
in ruminants given grass silage. Energy Metabolism.
(Ed. Mount, L.E.) pp, 357—362. Butterworths.

Tyrrel, H.F., Moe, P.W. & Bull, L.S. 1973. Energy
value of cracked corn and dried beet pulp. J. Dairy Sci.
56: 1384—1385. (Abstr).
& Reid, J.T. 1965. Prediction of energy value of

cow’s milk. J. Dairy Sci. 48: 1215—1223.
Uden, P. 1984. Digestibility and digesta retention in

dairy cows receiving hay or silage at varying concentrate
levels. Anim. Feed Sci. Technol. 11: 279—291.

—, Coi.ucci, P.E. & Van Soest, P.J. 1980. Investigation
of chromium, cerium and cobalt as markers in digesta.
Rate of passage studies. J. Sei. Food Agric. 31: 625
632.

Van Keulen, J. & Young, B.A. 1977. Evaluation of acid-
insoluble ash as a natural marker for digestibility
studies. J. Anim. Sci. 44: 282—287.

Visser, H., De & Groot, A.A.M., De. 1981. The in-
fluence of starch and sugar content of concentrates on
feed intake, rumen fluid, production and composition
of milk. Metabolic Disorders in Farm Animals (Ed. Gie-
secke, D., Dirksen, G & Stangassinger, M) pp. 41—48.
Veterinary Institute, Univ. of Munich.

Williams, C.H., David, D.J. & Riismaa, O. 1962. The
determination of chronic oxide in faeces samples by
atomic absorption spectrometry. J. agric. Sci., Camb.
59: 381—385.

Ms received April 27, 1987

SELOSTUS

Ohran, juurikaslcikkcen ja melassin vaikutus
maidontuotantoon, rehun sulavuuteen sekä
rehun viipymisaikaan ja virtausnopeuteen
vapaalla säilörehuruokinnalla

Pekka Huhtanen
Helsingin yliopisto, kolieläinlieleen laitos,
00710 Helsinki

Tutkimuksessa selvitettiin viiden erilaisen väkirehun vai-
kutusta maidontuotantoon, rehun sulavuuteen sekä vii-
pymisaikaan ja virtausnopeuteen vapaalla säilörehuruo-
kinnalla. Koe-eläiminä oli 10 Ay-lehmää, joista puolet oli
ensikkoja. Koe lehtiin kaksinkertaisena 5x5 latinalaise-
na neliönä. Väkirehut koostuivat seuraavista raaka-aineis-
ta (g/kg kuiva-ainetta (ka)): (A) ohra 1000 g, (B) ohra
667 g ja melassia 326 g, (C) ohra 417 g, leike 410 g ja
melassi 163 g,- (D) leike 983 g sekä (E) leike 656 g ja me-
lassi 326 g. Leikkeen ja melassin ohraa alempi raakaras-
vapitoisuus täydennettiin rypsiöljyllä. Eläimet saivat näitä
väkirehuja koko kokeen ajan 6 kg ka/pv, minkä lisäksi
annettiin I kg rypsirouhetta ja 250 g kivennäisseosta päi-
vässä.

Ohra- ja leikedieettien välillä ei ollut eroa säilörehun
syönnissä, mutta melassidieeteillä (BE) säilörehun ja sa-
malla kokonaiskuiva-aineen syönti vähenivät. Leikeruo-
kinnoilla (DE) maitotuotos oli 1.l kg (PcO.Ol) ja val-
kuaistuotos 32 g (P<0.05) korkeampi kuin vastaavat tuo-
tokset ohraruokinnoilla (AB). Sen sijaan maidon rasva-
pitoisuus oli leikedieeteillä merkitsevästi (P< 0.001) alempi
kuin ohradieeteillä (46.0 vs. 49.1 g/kg) ja siten rasvatuo-
toksessa ja rasvakorjatussa (RKM) maitotuotoksessa ei
ollut eroa näiden ruokintojen välillä. Ohran tai leikkeen

korvaaminen melassilla alensi maito- (P< 0.001), RKM-,
rasva- ja valkuaistuotosta (P < 0.01), mutta maidon koos-
tumukseen melassilla ei ollut vaikutusta. Parhaimman tuo-
toksen antoi kuitenkin monipuolinen rehuyhdistelmä (C),
joka sisälsi sekä ohraa, leikettä että melassia. Ero oli mer-
kitsevä maito-, RKM- ja rasvatuotoksessa (P<0.05) se-
kä valkuaistuotoksessa (PcO.Ol) neljän muun dieetin kes-
kiarvoon nähden.

Dieettien vaikutus rehun sulavuuteen määritettiin vii-
dellä lehmällä käyttäen suolahappoon liukenematonta
tuhkaa merkkiaineena. Dieettien orgaanisen aineen su-
lavuus oli keskimäärin 0.743, eikä dieettien välillä ollut
merkitseviä eroja. Raakavalkuaisen sulavuus oli ohra-
dieeteillä parempi (PcO.001) ja erilaisten kuitufraktioi-
den sulavuus huonompi (P< 0.001) kuin leikedieeteillä.
Laskettaessa säilörehun sulavuus erotuksella dieettien vä-
lillä ei kuitenkaan voitu todeta eroja.

Muuntokelpoisen energian hyväksikäyttö oli hieman
parempi leikedieeteillä kuin ohradieeteillä. Huolimatta
alentuneesta maidontuotannosta energian hyväksikäyttö
ei melassidieeteillä ollut huonompi, vaan huonompaan
tuotokseen oli syynä pääasiassa alentunut karkearehun
syönti ja osaksi suurempi elopainon lisäys.

Nestefaasin ja partikkeleiden virtausnopeus määritet-

119



tiin käyttämällä merkkiaineina CoEDTArta ja Cr-mer-
kittyä olkea. Nesteen keskimääräinen viipymisaika vaiti-
teli 18.0:sta 19.9 tuntiin ja partikkeleiden 36.8:sta 37.7
tuntiin dieetistä riippumatta. Partikkeleiden virtausnopeus
pötsistä vaihteli välillä 0.045—0.048 ja nesteen 0.096

0.104, eikä dieettien välillä ollut merkitseviä eroja.
Tämän sekä ulkomaisten tutkimusten perusteella leik-

keen ja ohran kuiva-aineen tuotantovaikutus lypsylehmillä
on sama huolimatta 12—15 %;n erosta niiden ry-arvossa
rehutaulukoissa.

120