 

 

 

”5" 34M.

«1.,

9’ {1.3.3.
:"ﬁ‘u' '
m

,f

a 4'1

w ‘ “A 'nﬂi‘zw'l'r

a “‘R‘ﬂ'; . ,l1gﬁ~.5.‘)r‘

1' g_ ~ Mr
..

...
; ‘_ ‘ . ‘ .—
a

. \ <-4-1 ‘ -‘ n if)‘
‘2 4:99:95? ‘5' ~“'* ‘7 3' 223% K
' “awn-4, r - ":r 2‘ =W-vm av‘ﬁ‘i' " .19.;
.5 . L; . as" 4.! . -1 “ﬁi'aw‘ ‘ '
s v i" k ﬁvﬁatv‘J-ﬂ‘gélﬂﬁ €53 . 2‘6 ant". WM'E‘HI
$11“. ’2‘; 5W " ' 9 . r."- -' ‘- ' ,Egd..‘m;#
. .- “‘ '1'.

.. a

"~de .

5“ .1 3321'”).
.

.1!
a;

9 5i
5.1“ ‘
“ ”£13“

a;
a?" 91%;? 3:9 .' "3
MW 25* -'.9:- v a:

a; . gﬁa$§§T§¢a

E. I

.35.
'3'

N. .

_ . . - '3
_,.v 221551 ~ ‘
#- ﬁy‘zégéglr ,. ”gin“; $112. gagiéuﬁm
, 1' '.
aiﬁhﬁa’éégﬁnii
: «.2299 a;
£3» »
gwﬁj.u¢g'ﬂ
. 4 £5“:

 

THE818

MICHKJAN STATE UNI RSITY LIBRARIES

llllllllll ll llllllllllllillllllllllll llllllll

3 1293 01033 2207

 

 

 

 

 

 

I".

This is to certify that the

thesis entitled

Effectiveness of Whole Fuzzy Cottonseed Neutral
Detergent Fiber Relative to Alfalfa Silage Neutral
Detergent Fiber at Two Theoretical Lengths of Cut

presented by

Charles Steven Mooney

has been accepted towards fulﬁllment
of the requirements for

Master's degree in Animal Science

 

Ma' professor

Date March 30. 1994

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

LIBRARY .
Mlchlaan State
Unlverslty

 

 

 

PLACE N RETURN BOX to remove this checkout from your record.
To AVOID FINES rotum on or More data duo.

DAngIaPE . DATE DUE DATE DUE

M
T”?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3
[l

‘7

 

 

 

i
I

 

 

 

 

 

3

M80 In An Affirmative Action/Equal Opportunity lmtltulon
M1

EFFECTIVENESS OF WHOLE FUZZY COTTONSEED NEUTRAL DETERGENT
FIBER RELATIVE TO ALFALFA SILAGE NEUTRAL DETERGENT FIBER AT
TWO THEORETICAL LENGTHS OF CUT.

By

Charles Steven Mooney

A Thesis

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Animal Science

1 994

ABSTRACT
EFFECTIVENESS OF WHOLE FUZZY COTTONSEED NEUTRAL DETERGENT
FIBER RELATIVE TO ALFALFA SILAGE NEUTRAL DETERGENT FIBER AT
TWO THEORETICAL LENGTHS OF CUT.
By

Charles Steven Mooney

Twelve Holstein cows (125 DIM) were fed diets differing in theoretical length of
cut of alfalfa silage (9.5 mm vs. 4.8 mm) and whole fuzzy cottonseed substituted for 0%
or 27% of alfalfa silage NDF. A 2 x 2 factorial arrangement of treatments was replicated
(n = 3) in a 4 x 4 Latin square design with 21 d periods. The objective was to determine
effectiveness of whole fuzzy cottonseed NDF relative to alfalfa NDF at two lengths of cut
as measured by the ability to stimulate chewing. Diets were formulated to 27% NDF.
Chewing activity was recorded with halter-mounted pressure sensors and a computerized
data collection system. Total chews per kilogram corrected (non grain) NDF intake were
decreased by cottonseed (P < .01) and short length of cut (P < .01) with an interaction
observed between the two effects (P < .05). Total chews per kilogram corrected NDF
intake attributed to whole cottonseed was divided by total chews per kilogram corrected
NDF intake attributed to alfalfa silage to calculate effectiveness. Effective NDF of
cottonseed was calculated as 48% and 80% relative to long and short cut alfalfa silage,

respectively.

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... vi
LIST OF FIGURES ........................................................................................................ viii
LIST OF ABBREVIATIONS AND SYMBOLS ........................................................... ix
REVIEW OF LITERATURE ......................................................................................... 1
Introduction ......................................................................................................... 1
Sequence of Events of Eating and Ruminating ................................................... 2
The Concept of Fiber Effectiveness .................................................................... 4
Inﬂuence of Particle Size .................................................................................... 6
Inﬂuence of NDF Content of the Diet ................................................................ 12
Inﬂuence of Maturity of Forage .......................................................................... 16
Inﬂuence of Animal Factors ............................................................................... 17
Measurement of Forage Fiber Effectiveness ...................................................... 19
Use of High Fiber By-products as Supplements ................................................. 23
Measurement of By-product Fiber Effectiveness ............................................... 25
Summary ............................................................................................................. 32
PURPOSE AND OBJECTIVES ..................................................................................... 34
MATERIALS AND METHODS .................................................................................... 35
Experimental Design ........................................................................................... 35
Cow Description ................................................................................................. 36
Cow Management ............................................................................................... 36
Forage Harvesting and Preservation ................................................................... 37

iii

iv

MATERIALS AND METHODS (continued) ................................................................ 35
Diet Formulation ................................................................................................. 37
Data Collected ..................................................................................................... 38
Samples Collected ............................................................................................... 39
Sample Analysis .................................................................................................. 39
Electronic Data Collection .................................................................................. 41
Statistical Analysis .............................................................................................. 42
Calculation of Fiber Effectiveness Coefﬁcient ................................................... 43
RESULTS ....................................................................................................................... 46
Comparison of Silages ........................................................................................ 46
Ingredient and Diet Composition ........................................................................ 48
Milk Production and Composition ...................................................................... 51
Feed Intake, Body Weight and Condition, and Cow Health ............................... 51
Chewing Activity ................................................................................................ 55
Drinking Activity ................................................................................................ 61
Correlations ......................................................................................................... 61
Results of eNDF Calculations ............................................................................. 64
Comparison of Responses by Square .................................................................. 65
DISCUSSION ................................................................................................................. 67
Ingredients and Diets .......................................................................................... 67
Cow Performance ................................................................................................ 69
Chewing Activity ................................................................................................ 70
Comparison of Responses by Square .................................................................. 71
Comparison of Chewing Activity in Other Experiments .................................... 72
eNDF Coefﬁcients for This Experiment ............................................................. 73
Comparisons to Effectiveness Coefﬁcients of Other Experiments ..................... 74

Comparison of Methods of Effective Fiber Calculation ..................................... 75

V

DISCUSSION (continued) .............................................................................................. 67
Assumptions of Calculations of eNDF for This Experiment .............................. 77
Choice of Method of eNDF Calculations for This Experiment .......................... 78

SUMMARY .................................................................................................................... 81

CONCLUSIONS ............................................................................................................. 82

FUTURE WORK ............................................................................................................ 83

APPENDICES ................................................................................................................ 85
Appendix A ......................................................................................................... 85
Appendix B ......................................................................................................... 86
Appendix C ......................................................................................................... 87
Appendix D ......................................................................................................... 88

LIST OF REFERENCES ................................................................................................ 89

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.
Table 6.
Table 7.

Table 8.

Table 9.

Table 10.
Table 11.
Table 12.

Table 13.
Table 14.
Table 15.
Table 16.

LIST OF TABLES

A sampling of roughage index values for cattle reported by

Sudweeks et al. (58). ............................................................................... 21
NDF percentages of high ﬁber by-product feed reported in review

by Harris, 1991 (33). ............................................................................... 24
Effective ﬁber coefficients for by-products reported by Swain et a1.

(61) and Vaughan et a1. (67). .................................................................. 30
Final treatment assignments for the 4 x 4 Latin square used in

experiment 92CSM1. .............................................................................. 36
Degrees of freedom for each source of variation. ................................... 43
Description of long and short cut alfalfa Silages. .................................... 47

Average percentage across periods of NDF residue retained on
standard sieves following wet sieving by length of cut. ......................... 48

Nutrient composition of forages and concentrates as measured
during the experiment for diet formulation and after experiment on
test week composites. .............................................................................. 49

Ingredient and nutrient composition of diets by treatment as
balanced during experiment and as measured postexperiment on

test week composites. .............................................................................. 50
Effect of treatment on milk production and composition means. ........... 52
Effect of treatment on intake, body weight, and body condition. ........... 53
Summary of animal health data during experiment listing the

number of occurrences during each treatment. ....................................... 54
Eating activity as affected by treatment. ................................................. 58
Ruminating activity as affected by treatment. ........................................ 59
Total activity as affected by treatment. ................................................... 60
Drinking activity as affected by treatment. . ............................................ 62

vi

Table 17.

Table 18.
Table 19.

Appendix A.

Appendix B.
Appendix C.

Appendix D.

vii

Pearson correlation coefﬁcients among 48 cow period means of

various feeding activity variables. .......................................................... 63
Fiber effectiveness coefﬁcients calculated in this experiment ................ 64
Comparison of responses by square. ....................................................... 66
Cow days lost for 92CSM] and reason they were removed from

ﬁnal data set. ........................................................................................... 85
Pre-experiment individual cow information for 92CSM1. ..................... 86
Summary of cow information by square on Sept. 8, 1992 for

92CSM1. ................................................................................................. 87
Formulas used in 92CSM1. ..................................................................... 88

LIST OF FIGURES

Figure 1. Total chewing time per kilogram corrected NDF intake reported in
studies with cattle varying particle size of alfalfa forage .............................. 11

Figure 2. The total chewing activity (min/d) at given corrected NDF intake in
cattle for several published studies reported in the Review of
Literature. ...................................................................................................... 15

Figure 3. Graphical representation of the formula used to partition the total
chews per kilogram corrected NDF intake. The calculation of ﬁber
effectiveness is Z divided by Y where Z = the proportion of chews
attributed of WFCS NDF = A-(l-B)(C), Y = the proportion of chews
attributed the replaced alfalfa silage NDF = (B)(C), A = total chews
per kilogram corrected NDF intake of diet containing WFCS, B = the
measured proportion of alfalfa silage NDF replaced by WFCS NDF,
and C = total chews per kilogram corrected NDF intake of diet without
WFCS. ........................................................................................................... 45

Figure 4. Graphical representation of assumptions and mathematics of this
study. The calculation of ﬁber effectiveness is the slope "B" divided
by slope "A." ................................................................................................. 76

Figure 5. Graphical representation of assumptions and mathematics of the slope
coefﬁcient method used by Armentano and coworkers (14, 15, 61, 67).
The calculation of ﬁber effectiveness is the slope "B" divided by slope
"A." ................................................................................................................ 77

viii

92CSM]
A:P

LIST OF ABBREVIATIONS AND SYMBOLS

code for experiment
acetate to propionate ratio
ad libitum

acid detergent

adjusted forage index

crude ﬁber

digestible ﬁber residue

dry matter

dry matter intake

electronic data collection
ether extract

effective neutral detergent ﬁber
hay crop silage

indigestible ﬁber residue
long cut, added WFCS diet
long cut, no WFCS diet
milliequivalents

Michigan State University
mean particle length

mean particle size

neutral detergent ﬁber intake
revolutions per minute
roughage value index

short cut, added WFCS diet
short cut, no WFCS diet
undegraded intake protein
whole fuzzy cottonseed

ix

REVIEW OF LITERATURE

Introduction

As ruminants, dairy cows need dietary ﬁber for proper digestive function. Proper
daily ﬁber intake is essential for optimum production and health of dairy cattle. Too
much ﬁber may limit high producing cows and too little ﬁber or ﬁber lacking coarse
texture may upset rumen fermentation, depress milk fat percentage, or threaten animal
health (46). Therefore, an optimum dietary ﬁber level exists for the high producing dairy
cow which will promote maximum energy and dry matter intake (46).

Determination of proper ﬁber intake requires accurate ﬁber characterization as
well as an accurate description of requirements. Currently, ﬁber is quantiﬁed by the Van
Soest detergent system (66) where the reagents are neutral and acid detergents. Neutral
detergent insoluble residue or ﬁber (NDF), corrected for ash content, contains mainly the
components of cell wall -- hemicellulose, cellulose and lignin. Acid detergent insoluble
residue or ﬁber (ADF), corrected for ash content, contains mainly cellulose and lignin.

While both ADF and NDF are used as measures of ﬁber, the residues of neutral
detergent most completely contain the components of cell wall and are, therefore, a more
accurate measure of total ﬁber (44). NDF is more closely associated with diet ﬁll as it is
positively related to bulk density (44), negatively correlated with DMI (44, 65), and
positively correlated with rumination time (71,73), while ADF is more negatively
associated with apparent digestibility (46). The ﬁber requirement for dairy cattle is better
expressed as NDF than ADF (66).

2

The amount of ﬁber recommended in a cow's diet is a function of many factors.
The National Research Council (NRC) subcommittee on dairy cattle research (46)
suggests the amount of ﬁber required by a cow is a function of such animal factors as
body condition, level of production, and feeding frequency, and such ﬁber factors as type,
physical character, amount, bulk density, and buffering capacity. For lactating cows,
NRC (46) recommends that the diet contain a minimum of 28% NDF and 75% of the
28% should be supplied by forage sources. An exception is made for cows producing
greater than 40 kg/day where the minimum diet NDF is lowered to 25%. No NDF
maximum is suggested -- only that ﬁber concentration may be increased when energy
needs are reduced.

As NRC (46) implies, ﬁber of different feedstuffs, particularly by-product
feedstuffs, may have unique chemical and physical properties. While the Van Soest
detergent system has advanced the measurement of ﬁber, the system is quantitative not
qualitative and, therefore, does not encompass all the properties and characteristics that

may be important to the cow.

Sequence of Events of Eating and Ruminating

As a ruminant, the cow reduces ﬁber particle size by chewing. The ruminant
chews for two proposes: eating and ruminating. Across a range of diets, a dairy cow will
spend 4 to 7 h each day eating and 5 to 9 h ruminating (8, 65). During this 9 to 16 h, the
cow chews 30,000 to 50,000 times (8, 65, 70). Ruminants tend to eat feed rapidly then
chew it more thoroughly later during rumination (8, 65). The chewing motion of
rumination is more uniform in rate and amplitude than eating (8, 65, 70).

Eating begins the digestive process and has three phases: prehension, chewing
and swallowing (8, 65). The tongue and lower incisors against upper dental pad gather
the feed and bring it into the mouth (8, 65, 70). The molars grind the feedstuff with a

lateral motion of the jaw reducing the feedstuff to a size that can be swallowed (70).

3
Coarser feedstuffs are chewed longer than ﬁne (8, 65). Saliva is mixed with the feedstuff
during chewing and the amount secreted is proportional to the time spent chewing (8, 65).
Once reduced in size, the feed is formed into a bolus and swallowed (8, 65, 70).

Rumination is the cyclic process of regurgitation, remastication, and reswallowin g
(65). Coarse ingesta stimulates tactile and pressure receptors located in the reticulorumen
(8, 65). The resulting nerve signals are sent to gastric centers in the medulla triggering
rumination (8, 65). A feedstuffs ability to stimulate these receptors is termed "scratch
factor." (8, 65) The amount of "scratch factor" in ingested particles is a function of
feedstuff particle size and total diet ﬁber content (8, 65) as well as forage species and
"toughness" as measured as resistance to digestion and physical reduction. Rumination is
stimulated by coarse ingesta particles (74), primarily by those particles longer than 10
mm (70), by increased particle toughness (20), and by high dietary cell wall content (71).
Rumination is inhibited by low pH, high osmotic pressure, or high volatile fatty acid
concentrations of ruminal ﬂuid (69).

Rumination begins with regurgitation of digesta (8, 65, 70). A bolus is formed by
reticular contraction, transported into the esophagus by negative pressure caused by
inspiration with the glottis closed, and moved to the mouth by rapid antiperistaltic motion
(8, 65, 70). Once in the mouth, regurgitated material is squeezed and excess liquid and
accompanying small particles are reswallowed (8, 65). The retained material is rechewed
and mixed with saliva for about a minute and then reswallowed (70). This cycle can be
repeated for some time and may be concurrent with other activities such as milking,
walking, urinating and defecating (8, 65). A cow will ruminate during 10 to 20 periods
per day with the total time of spent ruminating approaching a proposed upper limit of 10
hours (8).

Eating and ruminating stimulate saliva secretion (20). Saliva produced by the
cow is the predominant source of ruminal buffering (25). Saliva is produced by ﬁve sets

of paired glands and three unpaired glands (13). The parotid salivary glands account for

4
40 to 50% of the total saliva production (13). Saliva at a pH of 8.4 (25) contains mineral

ions (sodium, potassium, phosphate, bicarbonate), which provide buffering to the rumen

(8, 65). Saliva composition in cattle is fairly constant containing 125 meq/L HC03' and

26 meq/L HPO4' (25). Though measurement of saliva ﬂow in cattle has not been

perfected, mean daily saliva ﬂow is estimated at 171 L/d with a range of 108 to 308 L/d
(25). Therefore, given the composition of saliva, 390 to 1115 g/d of disodium phosphate
and 1134 to 3234 g/d of NaHCO3 are introduced to the rumen daily (25). If dietary
buffer were added at maximum of 1% of diet, the added acid neutralizing capacity of the
diet would still only be 10% of that provided by rumination (26).

Though there is always some ﬂow, chewing from eating and ruminating increases
saliva ﬂow (8, 65). Flows are estimated (12) to be 150 ml/min during resting, 177
m1/min during eating, and 300 ml/min during ruminating (These ﬂows yield the
maximum ﬂow of saliva reported in the review by Erdman (25)). Therefore, total daily
saliva ﬂow to the rumen is proportional to total time spent chewing (8, 65). Saliva ﬂow
may be inﬂuenced by forage intake (as forage intake increases, saliva ﬂow increases),
forage particle size (as forage particle size increases, saliva ﬂow increases), and the cow's

physiological state as lactating cows have a higher resting saliva ﬂow than dry cows (25).

The Concept of Fiber Effectiveness

In recent years, the concept of "effective ﬁber" has emerged. The concept is also
called "ﬁbrous characteristic" (6) and "roughage value" (45, 60). The concept is an
attempt to adjust the chemically deﬁned ﬁber fraction to better relate to responses by the
animal.

Effective ﬁber needs of a dairy cow are satisﬁed if ﬁber is fed in an adequate
amount of sirnulatory particle size (26). The minimum amount of effective ﬁber is
achieved if chewing activity and saliva production maintain the rumen pH above 6.2,

acetate to propionate ratio (A:P) above 2.2, and, consequently, maintain milk fat test

5
above 3.5 % (52). Lack of coarseness of forages and increased fermentation rate of diet
can have negative effects (8, 20, 26, 74) such as depressing feed intake below energy
needs, lowering milk fat content, and increasing the incidence of health disorders such as
acidosis, parakeratosis of rumen epithelium, bloat, displaced abomasum and laminitis.

Effectiveness of fiber is determined by the sum of many forage and animal
factors. Factors implicated in ﬁber effectiveness include animal species, individual
animal tendencies, animal body weight, animal physiological stage, time of access to
feed, dry matter intake, protein content of ration, forage to concentrate ratio of diet,
chemical composition of forage, particle size of the forage, type of concentrate, and
pattern of feeding (8, 20).

Fiber effectiveness has been measured by using total chewing time, ruminal
acetate to propionate ratio, or milk fat percentages. Higher chewing time, A:P, and milk
fat percentages are interpreted to mean increased effectiveness of ﬁber.

Most experimenters have used the resulting milk fat percentage on a given diet as
the measure of ﬁber effectiveness. Sufﬁcient effective ﬁber maintains a near neutral
ruminal pH (6.0 to 7.0) and promotes an acetate dominant fermentation in the rumen. An
acetate fermentation, in turn, increases fat percentage of milk produced. Milk fat is,
therefore, an indirect indicator.

Total chewing activity is a more direct measure of ﬁber effectiveness. Chewing
activity stimulates saliva flow and increased ﬂow carries more neutralizing ions to
counter the acids of rumen fermentation. Maintaining a near neutral ruminal pH helps
maintain rumen motility and dietary intake. Total chewing activity is also reﬂective of
the "scratch factor" maintaining the healthy rumen epithelium.

With all methods, effective NDF (eNDF) of a particular feedstuff is determined
by adjusting the chemical NDF measurement of cell wall content by a multiplier. The
multiplier is based on the ﬁber's relative ability to stimulate milk fat percentage or total

chewing activity. The multiplier may correct the chemical NDF measurement for such

6
factors as particle size, ability to stimulate rumination, fermentability and ﬁber source.
Implied in this correction is that the effectiveness of the NDF is measured relative to

something such as a coarse forage (e. g. long grass hay).

Inﬂuence of Particle Size

Effectiveness of ﬁber is inﬂuenced by particle size. Smaller forage particle size in
several forage species has reduced milk fat percentage (29, 30, 31, 53, 76, 77), ruminal
acetate to propionate ratio (30, 31, 53, 77), and chewing time (7, 21, 22, 29, 30, 31, 36,
40, 47, 48, 53, 74, 76, 77).

Mean particle size (MP8) or length (MPL) is determined as the weighted average
of material retained on sequential sieves. Forage material is sorted through a series of
sieves of decreasing aperture in geometric progression. The average obtained is not
really weighted by material length but by aperture opening. Several methods are used for
sifting material through sieves (shaking (dry) and ﬂushing with water (wet)) and for
calculating the variance of the distribution (log normal, gamma, exponential) around the
mean (2, 24).

Norgaard (47) fed three early lactation (42 DIM, 550 kg) cows three 20% barley
straw (78% NDF) and 80% concentrate diets in a 3 x 3 Latin square. The three diets were
generated by replacing 80%, 50% and 0% of the long barley straw with the same straw
pelleted. Chewing activity per day was determined using reticulorumen pressure
measurements. Eating activity was deﬁned as the presence of high frequency double
contractions and ruminating activity was deﬁned as the presence of triple contractions.
As fed intake was ﬁxed at 18 kg per day. Increasing the amount of the long barley straw
linearly increased time spent eating (from 83 to 104 min/d, P < .01), time spent
ruminating (from 125 to 403 min/d, P < .01), total chewing time (from 210 to 510 min/d,
P < .01), number of ruminating periods (from 8 to 17, P < .01), and ruminating bout

length (from 15 to 24 min/bout, P < .01).

7

For corn silage, reducing theoretical length of cut to 5 mm or less reduces
ruminating time and total chewing time. Okamoto et al. (48) fed three only corn silage
diets of three theoretical cuts: 4.8 mm, 9.5 mm, and 15.9 mm. The four cows on the
experiment were fed at 28 to 30 kg DM per cow per day (intakes not reported) and
chewing activity monitored with radio telemetry. NDF content of corn silage was not
reported but crude ﬁber was measured at 20% for all three Silages. Time spent
ruminating (min/d) was reduced (P < .05) with reduction in theoretical length of cut (576
min, 509 min, and 427 min on 15.9 mm, 9.5 mm and 4.8 mm, respectively). DeBoever et
al. (21) found that total chewing time per day increased with increased theoretical length
of cut (4 mm, 8 mm, and 16 mm) as a result of increased eating time per kilogram DMI
on the 16 mm cut (22 min vs. 20 min for 4 mm and 8 mm, P < .05) and decreased
ruminating time per kilogram DMI on the 4 mm cuts (36 min vs. 39 min on 8 mm and 16
mm, P < .05).

Reducing average particle size of alfalfa hay reduces chewing time (30, 36, 40,
53, 76), milk fat percentage (30, 53, 76), and ruminal acetate to propionate ratio (30, 53).
Kick et a1. (40) meal fed three steers only alfalfa hay of four particle sizes: an uncut
("long"), a theoretical length of cut of 2" ("coarse"), a theoretical length of cut 1/ "
(“ﬁne") and a hammer milled (" ground") and monitored eating and ruminating time per
day by counting chews with mechanical counters and noting the starting and stopping
times for each bout. The "ﬁne" and " ground" had less eating time ("ground" at 78 min/d,
"ﬁne" at 91 min/d, "coarse" at 130 min/d, "long" at 153 min/d, signiﬁcance not
determined) and "ground" had less ruminating time ("ground" at 277 min/d, "ﬁne" at 414
min/d, "coarse" at 403 min/d, "long" at 402 min/d, signiﬁcance not determined).

Shaver et a1. (53) fed early lactation (3 to 11 wk), mid lactation (20 to 32 wk), and
dry Holsteins diets containing 60% alfalfa hay of 41% NDF. The hay was three particle
sizes: uncut long, chopped to an average length of 7.8 mm and pelleted with an average

particle size of 1 mm (measured by dry sieving). When compared to long and chopped,

8
the pelleted hay depressed milk fat percentage (3.6% vs. 3.1%, P < .001), reduced
ruminal acetate to propionate ratio (3.9 vs. 2.5, P < .001), and reduced total chewing time
minutes per day (576 min vs. 189 min, P < .001) and minutes per kilogram DMI (29 min
vs. 10 min, P < .001). Particle size of hay had no effect on milk production (28 kg) or
milk protein percentage (3.12%). They concluded that long and chopped alfalfa hay were
equivalent and satisfactory in three measures of effectiveness.

Jaster and Murphy (36) fed 18 Holstein heifers (340 kg) three particle sizes of
63% NDF alfalfa hay in a 3 x 3 Latin square with consecutive 21 (1 periods. The three
particle sizes of hay were uncut long, coarsely chopped (average particle size was 2.2 mm
as measured by dry sieving), and ﬁnely chopped (average particle size was 1.4 mm as
measured by dry sieving). Total time spent chewing (16.7 h/d, P > .10) was not different
across diets. However, NDF intake was lower on long diets compared to coarse and ﬁne
diets (4.5 kg vs. 5.3 kg, P < .0001) which increased minutes of total chewing time per
kilogram NDF intake.

Woodford et a1. (76) fed four multiparous Holstein cows (56 DIM) in 4 x 4 Latin
square four 27% NDF diets of 47% alfalfa hay (47% NDF) which varied in measured
MPL. MPL of four chops of alfalfa hay were .26 cm, .46 cm, .64 cm, and .90 cm as
measured by dry sieving. Total chewing time (as recorded visually every 5 minutes for
24 hours) increased from 525 min/d on .26 cm hay diet to 604 min/d on .90 cm hay diet.
Milk fat percentage increased from 3.2% on .26 cm and .46 cm hay diets to 3.6% on .64
cm and .90 cm hay diets. Rurninal acetate to propionate ratio increased from 2.2 on .26
cm and .46 cm hay diets to 2.8 on .64 and .90 cm hay diets. However, these increases
were not signiﬁcant (P > .05). Four percent FCM production (27 kg), DMI (25 kg) and
chewing time per kilogram NDF intake (84 min) were not different (P > .05) across
treatments. They, none the less, concluded that milk fat depression was prevented when

forage MPL was 2 .64 cm.

9

Grant et al (30) fed alfalfa hay (46% NDF) at 55% of diet at three particle sizes to
four early lactation (21 DIM) multiparous Holsteins in a crossover design. Two chops, a
coarse and a ﬁne, dry sieved to an average particle size of 2.2 mm and 1.0 mm,
respectfully. The three diets were based on only coarse hay, only ﬁne hay, and a 50:50
mix of the ﬁne and the coarse (called "medium"). While NDF intake, DMI, and milk
production (kg) were similar across diets (6.6 kg, 22.7 kg and 25 kg, respectively, P >
.10), the ﬁne chop compared to medium and coarse had a lower milk fat percentage
(3.1% vs. 3.7%, P < .10). Reducing particle size lead to a reduction in time spent eating,
ruminating and total chewing per day (from 348 min/d on coarse to 322 min/d on ﬁne,
from 490 min/d on coarse to 381 min/d on ﬁne, and from 838 min/d on coarse to 704
min/d on ﬁne, respectively, P < .05). Also, with ﬁne, an increased ruminal propionate
(from 19 mol/100 mol on medium and coarse to 27 mol/100 mol on ﬁne, P < .05), a
decreased A:P (3.89 on coarse, 3.20 on medium, and 2.08 on ﬁne, P < .05), and a
decreased mean ruminal pH (6.25 on coarse vs. 5.80 on medium vs. 5.40 on ﬁne, P < .05)
were measured. Cows receiving the ﬁnely chopped diet also increased plasma glucose
and serum insulin. They concluded that measured blood and rumen factors were
supporting of the glucogenic theory of milk fat depression where increased ruminal
propionate shifts fat metabolism away from the mammary gland and to body adipose.

Grant et al (31) fed 18 early lactation (21 DIM) Holsteins three diets with alfalfa
silage (40% NDF) as 55% of the diet. The three particle size diets were generated with
two theoretical cuts: a coarse cut at .95 cm and a ﬁne cut at .48 cm. The three diets were
the coarse cut only, the ﬁne cut only, and a 50:50 mix of the two cuts. The MP8 of the
Silages were 3.1 mm, 2.0 mm and 2.6 mm, respectively, measured with dry sieving of
dried samples. Trends and changes were the same as the alfalfa hay study. Again, DMI
and NDF intake were similar across diets (22.2 kg and 6.3 kg, respectively, P > .05).
Milk production was similar for all diets (31.6 kg, P > .05), however, milk fat percentage

on the ﬁne diet (3.0%) was lower (P < .05) than the medium (3.6%) or coarse (3.8%)

10
diets leading to differences in 4% FCM (P < .05). Mean ruminal pH was lower (P < .10)
on the ﬁne diet (5.3) than medium (5.9) or coarse (6.0) diets. A:P increased (P < .05)
with increasing forage coarseness (2.77 on ﬁne, 3.13 on medium, and 3.52 on coarse).
Total daily chewing time in minutes (as recorded visually every ﬁve minutes for 24 h)
was increased with increased silage coarseness (570 min on ﬁne, 671 min on medium,
and 735 min on coarse, P < .001). In this study, the total chewing time in minutes per
kilogram DMI can be calculated as 25.4 min, 30.5 min and 33.1 min for ﬁne, medium
and coarse Silages, respectively. Total chewing time in minutes per kilogram corrected
(forage only) NDF intake was calculated by 91.0 min, 107.7 min and 116.5 min for ﬁne,
medium and coarse Silages, respectively.

Other studies using alfalfa silage of varied particle size (7, 29, 77) also showed
reduction in total chewing time (7, 29, 77) and milk fat percentage (29, 77) on ﬁner
forage particle size diets. DeBoever et al. (22) tried to vary grass silage particle sizes
with three cuts: an uncut, a 24 mm theoretical cut, and a 3.5 theoretical mm cut.
However, the measured average particle sizes were 48 mm, 24 mm, and 14 mm,
respectively, as measured by dry sieving. No difference was measured in total chewing
time per kilogram DMI (72.6 min, P > .05) across cuts but with the "short cut" averaging
14 m, it may not have been short enough to decrease total chewing time per DMI.

Figure 1 presents a summary of the inﬂuence of forage particle size on ﬁber
effectiveness as measured by total chewing activity. In those papers where both could be
determined, total chewing time in minutes per kilogram corrected (nongrain) NDF intake
is plotted against mean particle size in millimeters determined by dry sieving. In these
experiments (exception Woodford et a1. (7 6)), reducing dry sieve MPS reduced total

chewing time in minutes per kilogram corrected (nongrain) NDF intake.

ll

 

   

 

 

 

'0 250 i l l l l i
o
E) O
0
U A 200 T"- ——-
en on 0—0
"‘ i
,_ .
8N 150 -— --
E i
I- E f .
3%. 100 T” ‘—
.E Q
31:) Z _O_.Jaster and Murphy, 1983
U 50 __ + Shaver et al., 1986
.5 _9_Woodford et al., 1986
*5 _)(__ Grant et al., 1990b
[— —1— Grant et al.,1990a
0 l l i : l , ,
7 f
0 2 4 6 8 10 uncut -

Mean Aperture Size (mm)

Figure 1. Total chewing time per kilogram corrected NDF intake reported in studies with
cattle varying particle size of alfalfa forage.

12
Inﬂuence of NDF Content of the Diet

Increasing the NDF content of diets usually by increasing forage to concentrate
ratio has increased milk fat percentage (9, 76) and increased total chewing times (9, 10,
38, 57, 76). Reducing the diet NDF content by increasing the proportion of the diet from
grain leads to a more acid fermentation. A more acid fermentation can result in milk fat
depression. Increasing the proportion of the diet from grain also leads to a reduction in
total chewing activity on the diet. Decreased total chewing reduces the amount of saliva
produced and introduced into the rumen to neutralize the fermentation acids (8, 25, 65).
However, while total diet effectiveness was reduced in most experiments, effectiveness
per kilogram NDF was constant as measured by total chewing activity suggesting the
effectiveness of each unit of forage is constant.

Woodford et. al. (76) fed four multiparous Holstein cows (63 DIM) in a 4 x 4
Latin square and varied NDF content of alfalfa hay (46% NDF) based diets by altering
the forage to concentrate ratio. NDF contents achieved were 21%, 24%, 27%, and 30%.
On an average DMI across diets of 24.5 kg (P > .05), total chewing time and milk fat
percentage increased linearly with increasing NDF concentration. Total chewing time
increased from 629 min on 21% NDF to 794 min on 30% NDF but, total chewing time
per kilogram NDF intake was similar (115 min/kg NDF intake, P > .05) A:P increased
(P > .05) from 2.6 on 21% NDF to 4.0 on 30% NDF and milk fat percentage increased (P
> .05) from 3.5% to 4.0%. A numeric maximum 4% FCM of 32.3 kg was achieved at
27.5% NDF though 4% FCM output was similar on all diets (P > .05).

Jorgensen et a1. (38) fed 20 Holstein cows (49 DIM) in two groups. One group
was fed a diet that was 50% forage and the other diet was 33% forage on a DM basis.
The forage was alfalfa silage of 42% NDF. Fat corrected milk production was similar
across diets (27 kg/d, P > .05). Total chewing time per day (as recorded by visual
observation every 5 minutes for 72 hours) was decreased with decreasing forage (600 min

vs. 504 min, P < .05) but min per kilogram NDF were similar (102 min, P > .05).

13

Welch and Smith (71) reported a high correlation (r = .99) between rumination
time and NDF intake in sheep. They determined that " 1.05 min rumination time was
required for each gram of cell wall constituents." Welch and Smith (73) reported a
similar correlation for cattle (r = .94).

In 1977, Sudweeks (57) fed mature Jersey and Guernsey steers diets that were
90%, 60% and 30% forage. NDF contents were not reported. Combined results of the
factorial crossing the forage type (wheat silage, sorghum silage, or bermudagrass hay)
with concentrate type (citrus pulp, corn or soybean mill feed) showed an increase in
chewing time (P < .05) with increasing forage. Steers chewed 713 min/d on 90% diet,
490 min/d on 60% diet, and 387 min/d on 30% diet. Source of forage or concentrate did
not affect total chewing time (P > .05).

Beauchemin (9) fed six diets ad lib to six ruminally ﬁstulated Holsteins (80 DIM)
in a 6 x 6 Latin square with 28 d periods. The diets were composed of alfalfa hay and a
barley based concentrate. The hay and the concentrate were fed separately. The factorial
of diets crossed maturity of alfalfa hay (an early bloom of 39% NDF and a mid bloom of
48% NDF) and total NDF content of diet (31%, 34%, and 37%). DMI was similar across
diets (22.3 kg/d, P > .05) and total NDF intake (kg/d) was proportional to NDF content of
diet. Increasing NDF concentrations by adjusting forage to concentrate ratio tended to
linearly decrease milk production (from 27 kg to 25 kg, P = .09), linearly increased milk
fat percentage (from 2.7% to 3.3%, P < .001), but had no effect on FCM (22 kg, P =
.32). Increasing NDF concentrations linearly increased mean ruminal pH (from 5.6 to
6.0, P < .001) and linearly increased mean ruminal A:P (from 3.3 to 4.9, P < .001).
Chewing activities were monitored during the last week of each test period using a strain
gauge transducer built into a halter and linked to a computerized data acquisition system.
Increasing NDF of diet from 31% to 37% increased total chewing activity in minutes per
day (from 767 min to 852 min, P < .001), minutes per kilogram DMI (from 34.1 min/kg
DMI to 39.0 min/kg DMI, P < .001), but the diet did not effect minutes per kilogram

l4

NDF intake (106 min/kg NDF intake, P > .05) and decreased rrrinutes per kilogram hay
NDF intake (from 265 min to 144 min, P < .001). Increase of diet NDF content from
31% to 37% affected total chews similarly.

Figure 2 presents total chewing minutes per day for kilogram corrected NDF
intake in studies that reported both. Woodford et a1. (76) and Jorgensen et al. (38)
reported a linear relationship between total chewing time and kilogram corrected NDF
intake. Beauchemin (9) found a linear relationship between total chewing time and total
NDF intake, but not corrected NDF intake. Overall, the studies reviewed suggest a linear

relationship between total chewing time and kilogram corrected (nongrain) NDF intake.

 

 

 

 

1000 l l I 1
9.
g 900 .. r
E 800 a —
.?=‘
.g 700 — —
0
<1
.3) 600 — / _
B
o
S 500 — l—
0 —O- Woodford et al., 1986
T6 —0— Jorgenson et al., 1978 ﬁne
*5 400 -——0— Beauchemin et al., 199] early —
[- + Beauchemin et al., 1991 mid

-—t— Jorgenson et al., 1978 medium

300 I I I I
2 3 4 5 6 7

Corrected NDF Intake (kg/day)

Figure 2. The total chewing activity (min/d) at given corrected NDF intake in cattle for
several published studies reported in the Review of Literature.

16
Influence of Maturity of Forage

Effectiveness of ﬁber is inﬂuenced by maturity of forage. Most forages,
excluding corn, increase in NDF content as they mature. As the NDF content of a forage
increases and concurrently the digestibility of that NDF is reduced, the quality of the
forage is reduced as the whole plant becomes less digestible. Welch and Smith (73)
found a .94 correlation between rumination time and NDF content of test meals in dairy
heifers and steers. After fasting or feeding pelleted alfalfa for two days, a 5 kg (as fed)
meal of test forages varying in maturity was fed. Chewing behavior was recorded with an
air bladder on a halter and a paper recorder. As maturity of the forage increased (i.e.
NDF content increased and, though not reported, ﬁber digestibility probably also
decreased), rumination time was increased (P < .05). However, rumination time per gram
of NDF intake was constant (P > .05) at approximately .14 min/g NDF intake. Eating
times were not reported.

Beauchemin (9) fed an early bloom of 39% NDF and a mid bloom of 48% NDF.
Quality of hay had no effect on mean ruminal pH, mean ruminal acetate to propionate
ratio, milk production, or milk composition (P > .05). Chewing activities were monitored
during the last week of each test period using a strain gauge transducer built into a halter
and linked to a computerized data acquisition system. Total chewing activity (measured
either as minutes or chews) was affected by quality of hay (P < .05). Total chewing
activity as measured in minutes per day, minutes per kilogram DMI, and minutes per
kilogram NDF intake were lower on mid bloom alfalfa hay diets than early bloom hay
diets. However, when barley NDF was removed, total chewing activity per kilogram hay
NDF intake was higher on mid bloom hay diets than on early bloom (207 min vs. 188
min, P < .05) suggesting the more mature alfalfa hay of higher NDF and lower DM
digestibility was more effective.

DeBoever and coworkers have examined the effects of increased forage maturity

on chewing activity of 19 perennial ryegrass grass silages (2) and 14 com silages (21).

17
Eight cows were each fed a ﬁxed amount of supplement (3 kg for grass silage experiment
and 2.2 kg for corn silage experiment) and ad lib test forage. Chewing activity was
recorded by transducing air pressure changes in a halter mounted bladder to electric
signals and registering them with a paper recorder on the last four days of a test period.
Increased maturity of grass silage from "vegetative" to "full ears" increased total chewing
time per kilogram DMI (from approximately 80 min. to approximately 90 min, P < .05).
Increased maturity of corn silage from "ntilk dough" to "hard dough" decreased total
chewing time per kilogram DMI (from approximately 62 min to approximately 58 min, P
< .05). However, NDF content of forages were variable (mean 46.8% and range 41.7%
to 54.8% for grass silages and mean 38.5% range 35.7% to 43.2% for corn silage).
Generally, maturity increased NDF content of grass silages and decreased NDF content
of corn silages. Total chewing time per kilogram NDF was not reported for the grass
silages. Total chewing time per kilogram NDF was reported for the corn silages but, the

probability of maturity effects was not reported.

Inﬂuence of Animal Factors

The measurement of ﬁber effectiveness can be inﬂuenced by animal factors as
individual animals can respond differently to the same diets. This is particularly true of
chewing activities. Species of animal affects chewing activity over a range of diets.
Sheep and goats chew more per kilogram intake than cattle (20, 69). These differences
persist even when chewing activity is corrected for metabolic body weight (20, 69).
Sheep also chew faster than cows (20).

Within a species, individuals have distinct tendencies. In a review, DeBoever et
a1. (20) reported a coefﬁcient of variation of 15% for chewing activities per unit of DMI
among 14 dry pregnant British Friesian cows on a hay only diet. Dado and Allen (19)
reported daily chewing activity variables with coefﬁcients of variation as high as 22% for

12 Holstein cows (63 DIM) producing 21.7 to 47.6 kg/d milk per day and fed a common

l8
TMR of alfalfa silage, corn silage and concentrate. Coefﬁcients of variation in the
experiment (19) were 17.3%, 21.0%, and 20.1% for eating time (min) per day, per
kilogram DM, and per kilogram NDF, respectively. The ruminating coefﬁcients of
variation were 16.3%, 22.5%, and 22.3% time (min) per day, per kilogram DM, and per
kilogram NDF, respectively. And total chewing coefﬁcients of variation were 12.6%,
17.1%, and 17.0% for time (min) per day, per kilogram DM, and per kilogram NDF,
respectively, 1

Body size is a major source of chewing variation (70). Bae et al. (4) fed all hay
diets to 59 mature cows of several breeds and reported, as the body weight increased,
total chewing time per kilogram NDF decreases. Breed of dairy cow may also have
effects of chewing activity. Welch et al. (75) reported differences (P < .05) among dairy
breeds in rumination time in minutes per gram of NDF intake per kilogram metabolic
body weight. Guernsey had the lowest times then Holstein and Ayrshire and Jersey had
the highest times (5.3, 6.8, 7.0, and 9.2, respectively). Physiological stage of animal can
also inﬂuence chewing activity. In some experiments, pregnancy increased chewing time
per unit of feed (20).

As intake increases, eating, ruminating, and total daily chewing activities
increase. This increase in activity may (5) or may not (53, 59, 72) be constant chews per
kilogram intake. Bae et. a1. (5) fed four dry cows all hay diets in a 4 x 4 Latin square at
50%, 75% , 100% and 125% of NRC recommended intake. They found total chewing
time per kilogram NDF to be constant (about 135 min, P > .05). Welch and Smith (72)
fed 100% grass hay to sheep. As intake increased from 250 g to 1800 g, total rumination
time increased (from 231 min to 588 min), but rumination time per gram of NDF intake
decreased (from 1.36 min to .55 min, P not reported). Sudweeks et a1. (59) fed steers
90% corn silage diet (NDF content not reported) at three levels: 87% (7.4 kg), 122%
(10.4 kg), and 165% (12.2 kg) of "maintenance net energy milk"[sic]. Total chewing

times in minutes increased with intake (476 min, 628 min, and 706 min, P < .05) but,

19
total chewing minutes per kilogram DMI decreased with increased intake (67.5 min/kg
DMI, 62.8 min/kg DMI, and 57.1 min/kg DMI, P > .05).

To vary intake, Shaver et a1. (53) fed six Holstein cows a 60% alfalfa hay diets
during early lactation (3 to 11 wk), mid lactation (29 to 32 wk), and the dry period. The
prebloom alfalfa hay was obtained as bales from one ﬁeld and was 41% NDF. Dry
matter intakes were different (P < .001) at 24 kg for early lactation cows, 20 kg for mid
lactation cows, and 14 kg for dry cows. Total chewing activity (min/d) increased with
intake (386 min/d, 424 min/d, and 530 min/d, P < .001) and total chewing activity per
kilogram DMI was higher (P < .001) during dry period (27 min/kg DMI) than early or
mid lactation (21 min/kg DMI).

Measurement of Forage Fiber Effectiveness

Both chemical and physical characteristics of forage ﬁber affect their utilization
by ruminants and, therefore, both must be measured in an optimum eNDF system (45).
To measure the physical characteristics of a feed, Balch (6) proposed the use of "time
spent chewing per unit of DMI" as an index of "ﬁbrousness characteristic of roughages"
because physical character directly stimulates rumination.

Balch (6) selected daily total time spent chewing as part of his index because the
total time spent chewing was less variable than either time spent eating or ruminating.
Eating and ruminating time vary in opposition to each other (e. g. the more time spent
eating, the less time the cows spends ruminating) whereas total chewing time per
kilogram DMI relatively constant (20). Balch measured total chewing time by recording
pressure changes in halter mounted balloons as traces on a paper recorder. Total chewing
time was divided by kilogram DMI to correct for variation in amount consumed and feed
access time. Total chewing time is proportional to saliva production and easier to

measure (25).

20

Balch (6) reported indexes (total minutes chewing per kilogram DMI) obtained on
one ingredient diets fed to dry Short Horn and Friesian cows: oat straw was 145 to 191
min/kg DMI, "good quality" hay was 87 to 105 min/kg DMI, grass silage was 99 to 120
min/kg DMI, one ingredient pelleted concentrate diet was 4 to 29 min/kg DM1, and ﬁnely
ground hay diets was 13 to 19 min/kg DMI. The indexes of pelleted concentrate diet and
ﬁnely ground hay diet were a fraction of the coarse forages and were characterized by
irregular chewing activity.

Sudweeks and coworkers (58, 60) have tried to quantify effectiveness of forage
ﬁber through their "roughage value index system." Based on the proposal by Balch (6),
roughage value index (RVI) is deﬁned as the total time spent chewing per kilogram DMI.
They (60) fed 12 Holstein cows (70 DIM) alfalfa hay (all from one source) diets ad lib of
three forms: long, chopped at 3.81 cm theoretical cut, and pelleted. Grain was fed
separately at a constant amount to all cows within a block. Total chewing time per
kilogram DMI was measured with a bellows pneumograph on the halter and a paper
recorder system. They also fed these diets without any grain to six dry Holsteins.
Neutral detergent ﬁber contents were not reported. Mean total chewing time per
kilogram DMI for lactating cows was reported as 37 min on the long hay diet, 28 min on
the chopped hay diet and 15 min on the pelleted hay diet. The pelleted hay diet was
lower (P < .05) than the long or chopped hay diets. For dry cows, the indexes were 62
min, 44 min, and 37 min, respectively and the long hay diet was higher (P < .05) than the
chopped or pelleted hay diets. Regressing each lactating cow's milk fat percentage on her
minutes of chewing time per kilogram DMI, they reported that, to maintain 3.5% milk fat
or higher (their minimum criteria for ﬁber effectiveness), total diet index had to be 31
minutes per kilogram DMI or higher.

Sudweeks et a1. (58) also did an extensive review of the literature to collect and
calculate roughage index values. In diets that combined ﬁber sources, they used linear

regression to partition the total chew time per kilogram DMI to each ﬁber source. A

21
sampling of the values reported is in Table 1 (58). Using their literature review, they
developed a predictive equation that could be used to calculate individual feedstuff RVl
based on dry sieving particle size and chemical analysis as follows: Individual feedstuff
RVI (min/kg) = 10.86 «1» (dry sieve particle size in mean diameter)(21.59) — (DMI in
kg)(1.91) + (NDF content of feedstuff)(.541). Individual feedstuff RVI could be used to
calculate total diet RVI which they recommended should be at least 31 min/kg to
maintain a milk fat content of 3.5%.

Table 1. A sampling of roughage index values for cattle reported by Sudweeks et a].
(58).

 

Total Chewing Time/kg DM, min.

 

Alfalfa hay, long 61.5

Alfalfa hay, pelleted 36.9

Oat straw 160.0
Orchard grass, early cut 74.0
Orchard grass, late cut 90.0

Alfalfa silage 22.3 to 26.0
Corn silage 40.0 to 66.1
Grass silage 90.0 to 120.0
Barley ground 15.0

Ground corn 1.0 to 9.2

 

Santini et al. (50) proposed another system that would combine chemical and
physical measurements in an effort to measure diet ﬁber effectiveness. They proposed an
adjusted forage intake system which was calculated by multiplying forage intake in
kilogram by MPL. MPL is determined as the weighted average of material retained on
standard sieves after dry sieving. They hypothesized that, if adjusted forage intake (AFI)
correlated to total chewing time, the tedious measurement of chewing activities could be
avoided. They fed six multiparous Holsteins (60 DIM) in two 3 x 3 Latin squares alfalfa
based diets ad lib. One Latin square had 33:67 forage to concentrate ratio and the other

was 50:50. Forage component of diet differed (P not reported) in mean particle length:

22
.31 cm, .43 cm, and .51 cm. The differing diet particle size were achieved by substituting
alfalfa hay (55% NDF, 15% CP) for alfalfa hay crop silage (45% NDF, 21% CP),
confounding particle size and ﬁber content. The .31 cm diet was achieved with 100:0
silage DM: hay DM ratio, .43 cm with a 67:33 and .51 cm with a 33:67. They monitored
chewing activity visually every ﬁve minutes for 24 h and calculated AFI in kg-cm. They
reported that the relationship between total chewing time in min per kg-cm (Y) and
adjusted forage intake in kg-cm (X) was Y = 365.2 - 69.6 X + 4.73 X2. As adjusted
forage intake (kg'cm) increased to 125, total chewing time in min per kg~cm decreased
until AFI = 7 where became asymptotic to X axis. They concluded that adjusted forage
intake was closely associated with total chewing time over the range of forage intakes
(7.7 to 12.2 kg/d) and mean particle length (.31 to .51 cm) in their study and that the use
of AFI to calculate the roughage index of forages had merit.

DeBoever and coworkers also tried to replace the measurement of total chewing
index with a combination of chemical and physical lab methods. They fed eight cows in
two experiments (described in an earlier section): one for grass silage (22) and one for
corn silage (21). They varied maturity, length of chop and year of cutting. They
measured chewing activities electronically and reported a range of indexes for grass
silage of 58.5 to 96.9 min per kg DMI and for corn silage of 50.8 to 75.5 min per kg
DMI. They also. measured dry matter content, crude ﬁber content, NDF content, ADF
content, lignin content, sugar content, dry sieve particle distributions, milling resistance,
forage density and in vitro digestibility and regressed them on total chewing indexes of
the silages. For their grass silages, crude ﬁber content accounted for 74% of variation in
total chewing index and the addition of physical characteristics to the regression added
little. However, the shortest measured particle size was measured at greater than 14 mm.
For corn silage, in vitro digestibility combined with percent of non grain DM retained by
sieves of 22.38 mm (they considered this material stimulatory to rumination) accounted

for 87% of variation. In vitro digestibility alone accounted for 66% of the variation and

23
percent of non grain DM retained by sieves of 22.38 mm accounted for 63% of the

variation.

Use of High Fiber By-products as Supplements

The ﬁber in a dairy cow's ration normally comes from on-farm forages. On-farm
forage NDF can be supplemented at luxury or at necessity with high NDF by-products.
High ﬁber by-products are often an inexpensive source of energy and bypass protein.
However, while some by-products may be similar to forages in NDF content, by-product
NDF may have a diminished ability to stimulate chewing, to maintain near normal
ruminal pH and to maintain milk fat content of approximately 3.5% (26, 56).

By-products that have been used to supplement ﬁber are listed in Table 2.
Corrugated paper and cottonseed hulls are almost all ﬁber while others (distiller's grain
and whole fuzzy cottonseed) add ﬁber as part of a total package of ﬁber, energy, and
protein. Some are highly digestible (beet pulp, pineapple bran, soybean hulls) and others
(rice hulls, sunﬂower hulls, oat hulls) have low digestibility (33,35). Whole fuzzy
cottonseed and soybean hulls are used extensively throughout the United States while
peanut hulls, beet pulp, brewers grains, and distillers grains are regionally available (33,
56). By-products are highly variable in ﬁber composition and digestibility (56).
Consequently, estimates of effectiveness within a by-product can be highly variable,
therefore, measurements, where possible, are recommended (56). The by-product feed

variability of chews per kilogram NDF intake is not known.

24

Table 2. NDF percentages of high ﬁber by-product feed reported in review by Harris,
1991 (33).

 

 

% NDF of DM
Beet pulp 54
Citrus pulp 23
Corn cobs 85
Corrugated paper 92
Cottonseed hulls 90
Dried brewer's grains 46
Oat hulls 78
Peanut hulls 74
Pineapple bran 73
Rice hulls 80
Soybean hulls 67
Sugarcane bagasse 86
Sunﬂower hulls 83
Whole fuzzy cottonseed 44

 

Whole fuzzy cottonseed (W FCS, also called "whole cottonseed with lint"; IFN 5-
01-614) is an attractive by-product for feeding lactating dairy cows as WFCS is high in
ﬁber, energy, and protein (particularly UIP). These multiple beneﬁts often make it part
of the diet in a linear programming or a least cost environment (16). NRC (46) lists
WFCS as 2.23 Mcal NEng, 23% crude protein, 20% ether extract, 44% NDF, 34%
ADF, and 5% ash.

Whole fuzzy cottonseed has feeding limits because it contains gossypol.
Gossypol, a phenolic compound found in the pigment glands of cottonseed, is toxic to
animals (16). However, ruminants are less affected than nonruminants because gossypol
is bound in rumen (16). In ruminants, gossypol toxicity is possible but unlikely in typical
feeding rates of 3 to 4 kg per day (16). Symptoms of toxicity are depressions in
hemoglobin and total blood proteins and increases erythrocyte fragility (16, 17). Whole
cottonseed can be fed to lactating dairy cows at up to 25% the diet with no effect on total

intake (16, 56).

25
When fed to lactating cows, WFCS generally tends to increase milk yield and

milk fat test while decreasing milk protein test (16). However, results may depend on
forage base of diet (55). WFCS addition to corn silage based diets at 10% to 15% tends
to increase milk production and decrease milk fat (56, 64) and WFCS addition to alfalfa
based diets at 10% to 30% tends to maintain milk yield but increase milk fat percentage
(54). Also, the addition of WFCS to lactating dairy diets can affect milk fat production
by altering the fatty acid composition of the milk fat. Whole cottonseed in diets reduces
the proportion of short chain (C6 to C14) fatty acids in milk fat and increases the dietary

transfer of longer chain (C18) fatty acids to milk fat. (34, 54, 55)

Measurement of By-product Fiber Effectiveness

Since by-products can not generally be fed at 100% of diet, by-products must be
fed as a portion of the diet The challenge then becomes separating the by-product effects
from the other diet components.

In one of the earlier studies of by-product effectiveness, Lofgren and Warner (42)
measured by-product effectiveness using milk fat percentage response. They compared
beet pulp, corn cobs, and cat hulls to a mixed grass/legume hay. After a minimum ﬁve
week preliminary period in which a 20% mixed hay and 80% pelleted concentrate diet
was fed to depress milk fat percentage, 21 lactating Holsteins (270 DIM) were fed one of
four test diets for 16 weeks. These diets were all 14% crude protein and 15% crude ﬁber
with 60% of the crude ﬁber coming from hay (38% crude ﬁber, 68% NDF and 41 %
ADF). The remaining 40% of crude ﬁber came from one of four sources: beet pulp
(21% crude ﬁber, 40% NDF and 22% ADF), corn cobs (28% crude ﬁber, 69% NDF and
33% ADF), oat hulls (33% crude ﬁber, 75% NDF and 37% ADF), or additional hay as a
control. (All component concentrations are reported on air-dry basis.) After two weeks
on experiment diets, monitoring of milk production and composition began. All

experiment diets corrected the milk fat depression (2.3% vs. 3.3%, P < .001) induced by

26
the pre-experiment diet. No milk fat differences across treatment diets were found
suggesting that by-product crude ﬁber is just as effective as added hay crude ﬁber as
measured with milk fat percentage. Using data from each individual cow, they correlated
fat test change (from pre—experiment to experiment) against change in percent intake of
dietary components from pre-experiment to experimental diets. Change in percent intake
of ADF was the most highly (P < .05) correlated with change in fat test. Change in
percent intake of crude ﬁber and change in percent intake of cellulose were correlated (P
< .05), but to lesser extent. Change in percent intake of NDF was not correlated (P >
.05).

In 1991, Beauchemin et al. (11) reported an experiment to examine the ﬁber
effectiveness of pelleted alfalfa, beet pulp and oat hulls relative to alfalfa hay. Nine
lactating Holsteins (80 DIM) in two 5 x 5 Latin squares were fed ﬁve alfalfa hay forage
(42% NDF) and barley based concentrate diets. Hay and concentrate fed sequentially.
The ﬁve diets were: a 29% NDF diet with 65% of dry matter from the concentrate, a 32%
NDF diet with 65% of dry matter from the concentrate which contained added suncured
alfalfa pellets, a 32% NDF diet with 65% of dry matter from the concentrate which
contained added beet pulp, a 32% NDF diet with 65% of dry matter from the concentrate
which contained added oat hulls, and a 32% NDF diet with 55% of dry matter from the
barley based concentrate. They recorded chewing time and number using a strain gauge
transducer mounted on a halter and linked to a computerized data acquisition system,
milk production and composition, and rumen parameters. DMI was similar for all diets
(21 kg/d), but NDF intake was higher on 32% NDF diets than the 29% NDF diet (6.7
kg/d vs. 6.1 kg/d, P < .001). Total chewing activities were similar across diets in per day
measures (740 min/d and 46,000 chews/d, P > .05) and in per kilogram DMI measures
(35.5 minutes per kg DMI and 2220 chews per kg DMI, P > .05). However, total
chewing activities per kg NDF intake were greater on the 29% NDF diet than on 32%
NDF diets (126 min and 7940 chews vs. 111 min and 6915 chews, P < .05). Milk

27
production was similar across diets (26 kg/d, P > .05). Milk fat percentage was lowest on
the 29% NDF (2.2%), intermediate on NDF supplemented concentrate diets (2.9%) and
highest on the 32% NDF all hay diet (3.2%, P < .05). Fat corrected milk was greater on
32% NDF all hay diet as well (23.2 kg/d vs. 21.5 kg/d, P <.05). Mean ruminal acetate to
propionate was lowest on the 29% NDF (1.57), intermediate on NDF supplemented
concentrate diets (1.94), and highest (2.27)on the 32% NDF all hay diets (P < .001).
They concluded when forage must be limited in the diet, suncured alfalfa pellets, beet
pulp, or oat hulls can be added to enhance milk fat percentage. However, increasing NDF
content of the diet by increasing the forage to concentrate ratio of the diet is more
effective than by-product supplementation. Thus, the effectiveness of NDF of suncured
alfalfa pellets, beet pulp and oat hulls is greater than 0% but less 100% when compared
with alfalfa hay.

Work from The Ohio State University reported in 1991 (51) and in 1992 (27)
tested the validity of the NRC (46) recommendation that 75% of total diet NDF should
come from forage. Sarwar et al. (52) fed ﬁve duodenally and ruminally cannulated
Holstein heifers (500 kg) in a 5 x 5 Latin square with 17 (1 periods. The ﬁve diets all at
31% NDF were based on 50:50 (DM basis) corn silage to alfalfa. Using soyhull and corn
gluten feed NDF to replace forage NDF, the ﬁve diets were: a control diet with 85% of
the diet NDF from the forage mix; a test diet with 65% of the diet NDF from forage and
20% from corn gluten feed, a test diet with 45% of the diet NDF from forage and 40%
from corn gluten feed; a test diet with 65% of the diet NDF from forage and 20% from
soyhulls; and a test diet with 45% of the diet NDF from forage and 40% from soyhulls.

A tendency toward lower ruminal pH 3 h post feeding was recorded on 45% forage NDF
diets when compared with 65% forage NDF diets (5.6 vs. 5.9, P = .07). Ruminal A:P
was similar across diets (P > .05) at 3 h (3.5), 6 h (3.3), and 9 h (3.7) post feeding. Based
on ruminal and whole tract measurements, they (51) suggested that 65% of dietary NDF

from forage may be sufﬁcient when corn gluten feed or soyhulls replace the forage.

28

In a follow up study, Firkins and Eastridge (27) replaced forage NDF with soyhull
NDF in lactating cow diets. After common pretreatment diet, 40 early lactation (4 weeks)
cows were blocked for parity, previous milk and DIM. These cows were fed 28% NDF
diets in a randomized block design for 16 weeks. A control diet contained 75% of NDF
from a forage mix of 3:1 (DM basis) corn silage (N DF content not reported) and alfalfa
silage (55 to 58% NDF). Two of the three test diets had soyhulls (NDF content reported)
replacing corn silage on a DM for DM basis to lower the forage contribution of the diet
NDF to 62.5%. These three diets measured 36% NDF. On these three diets, intakes (24
kg DM/d), body weight change (.36 kg/d), 4% FCM (33 kg/d), milk fat percentage
(3.4%), and milk protein percentage (3.2%) were not affected by diet (P > .10). Even
though these diet NDF contents were higher than the NRC (46) recommendations and
77% of the required 28% NDF was form forage (.60 x 36%+ 28%), they concluded the
results suggest the addition of soyhulls can reduce forage contribution to total NDF to 60
to 65%.

Wagner et. a1. (68) replaced corn silage (NDF content not reported) with wheat
middlings (% NDF not reported). Five lactating cows in 5 x 5 Latin square were fed ﬁve
diets which were 33.5% NDF. The high forage control had 50% of dry matter as corn
silage. Four test diets had 15% of diet as wheat middlings replacing corn silage. The
addition of wheat middlings to replace corn silage increased DMI (19.3 kg vs. 20.6 kg, P
= .01) and had no effects (P > .10) on milk production (30 kg/d), milk composition
(3.6% milk fat, 3.5% milk protein), or ruminal pH or acetate to propionate ratio. They
concluded that wheat middlings at 15% of dry matter can effectively substitute for corn
silage in lactating cows diets. This suggests wheat middlings equal corn silage in ﬁber
effectiveness.

Research at the University of Wisconsin is the ﬁrst attempt to quantify
effectiveness of by-products. A series of studies reported in the 1990's(14, 15, 61, 67)

examined a variety of by-products. In these studies, by-product NDF effectiveness was

29
calculated relative to alfalfa silage NDF using milk fat percentages. To establish the
standard in each of these experiments, milk fat percentages were measured on a low and
high NDF control diets. The increases in milk fat percentage from low NDF to high NDF
alfalfa silage diets were divided by the increase in the nongrain NDF content of total diet
from low to high. This ratio was called the "slope coefﬁcient." By-product NDF was
also added to the basal low alfalfa silage NDF diet and slope coefﬁcients for these by-
products were calculated comparing the high NDF from by-product diet to the basal NDF
diet. The slope coefﬁcients from various by-products were divided by the slope
coefﬁcient established for alfalfa silage on a given experiment to give the relative
effectiveness of by-product NDF compared to alfalfa silage NDF.

Swain et al. (61) reported NDF effectiveness using the slope coefﬁcient method
for brewer's grain, oat hulls, corn gluten feed, beet pulp and malt sprouts. Using two
experiments, they fed 21 multiparous midlactation Holsteins (DIM not reported) in an
incomplete block design with seven treatment diets in three 21 (1 periods. With
concentrate NDF negated, the low NDF alfalfa silage diet was 14.4% nongrain NDF and
the high NDF alfalfa silage diet was approximately 21% nongrain NDF. For test diets,
they added by-product NDF to basal diet to bring total diet nongrain NDF content to the
same level as the high alfalfa silage control. Slope coefﬁcients were calculated for each
by-product and alfalfa silage. By-product ﬁber effectiveness was approximately 50% of
alfalfa silage. (Actual coefﬁcients are presented in Table 3.) Milk fat test on low NDF
control was 2.5% and on high alfalfa silage control was 3.0%. They reported that added
brewer's grain, oat hulls and corn gluten feed raised fat test above basal diet (P < .01) but

not to the extent of high alfalfa diets (P < .10).

30

Table 3. Effective ﬁber coefﬁcients for by-products reported by Swain et al. (61) and
Vaughan et al. (67).

 

 

Coefﬁcient Source
Beet pulp .44 (61)
Brewer grains .33 (6])
Corn gluten feed .56 (61)
Ground corn cobs .45 (67)
Malt sprouts .47 (61)
Oat hulls .66 (61)
Wheat middlings .57 (67)

 

Vaughan et al. (67) compared the effectiveness of NDF from ground corn cobs
and wheat middlings against alfalfa silage using the slope coefﬁcient method. They used
three 4 x 4 Latin squares with 21 day periods and 12 multiparous midlactation (DIM not
reported) cows. Again, negating concentrate NDF, the four alfalfa silage diets were: (1)
the basal diet at 23% nongrain NDF, (2) the added alfalfa diet (high control) at 27.4%
nongrain NDF, (3) the added ground corn cobs diet at 29% nongrain NDF, and (4) added
wheat middlings diet at 28% nongrain NDF. DMI (25 kg/d) and milk yield (32 kg/d)
were not different (P > .05) across diets. Milk fat percentages were 3.05%, 3.38%,
3.24%, and 3.27% for diets (l), (2), (3), and (4), respectively. To contrast milk fat tests,
three comparisons were used: (3) vs. (4), (3) and (4) vs. (1), and (3) and (4) vs. (2). Only
the (3) and (4) vs. (1) contrast was reported to be signiﬁcant (P < .05). Again, slope
coefﬁcients were determined and effectiveness of NDF relative to alfalfa silage was
calculated to be approximately half. (Actual coefﬁcients are presented in Table 3.)

In 1993, Clark and Armentano (15) reported determinations of effectiveness of
NDF using the slope coefﬁcient method for dried distiller's grain (32% NDF) and whole
cottonseed (45% NDF) relative to alfalfa silage (43% NDF, MPL = 7.6 mm by dry
sieving of wet material). They fed sixteen multiparous Holstein (135 DIM) cows in 4 x 4

Latin squares with 21 (1 periods. The four diets were: (1) a basal diet 13% NDF with all

31
NDF from alfalfa silage, (2) a high control alfalfa silage diet at 19% NDF, (3) the basal
diet plus dried distiller's grain NDF equaling 6% of diet (total diet = 19% nongrain NDF),
and (4) the basal diet plus whole cottonseed NDF equaling 6% of diet (total diet = 19%
nongrain NDF). Calculated diet NEL were 1.77 Mcal/kg for (l), 1.71 Mcal/kg for (2),
1.75 Mcal/kg for (3), and 1.78 Mcal/kg for (4). Contribution of concentrate NDF was
negated in diet NDF content calculations. The fat contents (as measured as the sum C14
to C18 fatty acids) of diets were reported as 5.0% for low ﬁber control, 4.7% for high
ﬁber control and dried distiller's grain added diets and 4.2% for whole cottonseed added
diets (standard error not reported). The high control alfalfa diet had lower (P < .05) DMI
(22.8 kg vs. 24.0 kg), milk production (30.6 kg vs. 32.1 kg) and FCM (27.1 kg vs. 28.2
kg) than other three diets. Milk fat percentages were 3.16%, 3.30%, 3.27%, and 3.34%
for the basal diet, high control alfalfa diet, added dried distiller's grain ﬁber and added
whole cottonseed ﬁber diets, respectively, and were not different (P > .10). Ruminal
contents were collected by stomach tube once for VFA analysis. Ruminal A:P obtained
was higher (P < .01) on high control alfalfa diet than other diets (3.7 vs. 2.9). Total
chewing activity (measured visually every ﬁve minutes for one 24 h period) in minutes
per day was higher (P < .05) on (2) and (4) than on (1) and (3) (754 min/d vs. 666 min/d).
From the numbers reported, total chewing time minutes per kilogram nongrain NDF
intake was calculated as 220 min/kg, 176 min/kg, 155 min/kg, and 167 min/kg for (l),
(2), (3), and (4), respectively. Using the slope coefﬁcients method, effectiveness of NDF
was calculated as .80 for dried distiller's grain and 1.30 for whole cottonseed relative to
1.0 for alfalfa silage. Clark and Armentano concluded the NDF of the by-products was
approximately as effective as alfalfa silage in stimulating fat synthesis. They also stated
the nonsigniﬁcant differences of fat tests lowered the precision but not accuracy of
calculated effectiveness factors. However, they cautioned that corroborating studies were
needed before the NDF effectiveness factor of 1.3 for whole cottonseed could be

recommended for use.

32

In a follow up study, Clark and Armentano (14) tested the validity of the
coefﬁcients they derived. Using a 4 x 4 Latin square design, they fed 16 multiparous
Holsteins (DIM not reported) four diets in 21 (1 periods. Using slope coefﬁcient methods
for comparison of diet, they fed a basal diet at 12% nongrain NDF and a high ﬁber
control diet at 22% nongrain NDF with the increase achieved by the addition of alfalfa
silage NDF. In contrast to previous work, forage NDF of basal proportion of the diet was
not solely from alfalfa silage but from a mix contributing equal amounts of NDF from
corn silage and alfalfa silage (NDF contents of forages not reported). The two test diets
added a mix of nonforage ﬁber (dried distiller's grain, whole cottonseed and wheat
middlings; mix proportions and NDF contents not reported) to basal forage to achieve
total diet nongrain NDF contents of 17% (NFFl) and 22% (NFF2). Based on previous
work, they expected the NFFl diet to have a milk fat response midway between basal and
high ﬁber diets and the NFF2 diet to have a milk fat response equal to the high ﬁber diet.
DMI (22 kg) and milk production (33 kg) was similar on all diets (P > .05) but milk fat
and milk protein percentages were affected by treatment (P < .05). Contrary to
prediction, milk fat percentage on NFF2 diet (3.2%) was not equivalent to milk fat
percentage of standard diet (3.5%) and milk fat percentage of NFFl (2.9%) was not
midway between standard (3.5%) and basal (2.8%) diets. These results suggest the
effectiveness coefﬁcients determined in previous experiments using milk fat percentage
were overestimated or that the change in forage base had an effect. Chewing time
(min/d) was affected by treatment (P < .05) with standard diet (457 min) higher than
basal, NFFI, and NFF2 diets (394 min, 375 min, and 409 min, respectively).

Summary
The concept of effective ﬁber is an attempt to quantify the properties of feedstuff
ﬁber beyond the chemical measurement of ﬁber. Fiber is effective when it stimulates

chewing and salivation for increased milk fat by increased acetate production.

33
Effectiveness of ﬁber is determined by the sum of many factors. Reduced forage particle
size by chopping too ﬁne or pelleting reduces ﬁber effectiveness as measured as total
chewing per kilogram NDF intake. Lowering NDF content of diet reduces the
effectiveness of the diet as measured by total daily chewing activity, but effectiveness of
forage as measured by total chewing activity per kilogram remains constant. Increasing
maturity of forage tends to increase effectiveness as measured as total chewing activity
per kilogram NDF intake. Fiber effectiveness measures may be inﬂuenced by such
animal factors as species, body size, and intake.

The index of ﬁber‘s physical character has been recommended to be time spent
chewing per unit of DMI (6). Several effective ﬁber systems have tried to use this or a
correlated substitute based on chemical, physical or production measurements to develop
practical ration balancing systems. The use of activity per unit of DMI is a concern
because it does not account for the great variation of ﬁber content across the range of
feedstuffs used.

High ﬁber by-products can be used to supplement diet ﬁber. WFCS is an
attractive ﬁber supplement because it is also high in protein and energy. By-product
NDF effectiveness has been measured by chewing activity, ruminal A:P, or, most
commonly, by its ability to correct milk fat depression. By-product NDF is generally less
effective than coarse forage NDF but it depends on source of the by-product ﬁber. The
range of effectiveness coefﬁcients have been determined from .33 for dried brewer's

grains (61) to 1.3 for WFCS (15).

PURPOSE AND OBJECTIVES

WFCS is widely used in Michigan as dairy feed. Knowing the effectiveness of
WFCS NDF is important because the high producing cows that receive WFCS are
approaching the lower limits of dietary ﬁber where accurate ﬁber balance is important for
maximizing energy intake.

The objective of this study was to measure the effectiveness of WFCS NDF
relative to alfalfa silage at two lengths of cut using total chews per kilogram NDF intake
to calculate effectiveness. Chewing activity was viewed as a direct indicator of NDF
effectiveness as the stimulation of chewing activity increases salivation to neutralize the
fermentation acids in the rumen. The null hypothesis postulated that WFCS NDF was
equal to alfalfa silage NDF in its ability to stimulate chewing activity.

The sub objectives of this experiment were:

1. To compare the number and type of chews from WFCS NDF and alfalfa
silage NDF.

2. If differences existed, to examine chewing activity as affected by length
of cut by NDF source interactions.

3. To calculate the effective ﬁber coefﬁcient of whole fuzzy cottonseed
relative to alfalfa silage of two chop lengths based on total chew counts divided by

intake.

34

MATERIALS AND METHODS

Experimental Design

An experiment designed as a 4 x 4 Latin square was replicated (n=3) and balanced for
carryover effects. Three squares containing twelve midlactation (125 DIM) cows were
run concurrently with four consecutive 21 (1 periods. The four treatments were generated
by a 2 x 2 factorial. The factorial design crossed alfalfa silage length of cut ("long" and
"short") with WFCS treatment. The long cut alfalfa was obtained with a forage harvester
set at a 3/8" (9.5 mm) theoretical length of cut and was typical of material removed from
a bunker silo. The short cut alfalfa was a 3/ 16" (4.8 mm) theoretical cut and was typical
of material removed from a bottom unloading upright silo. Treatment with WFCS called
for the replacement of 0% or 25% of the forage NDF with WFCS NDF. Each period
began with a 14 day adjustment period and ended with a seven day collection period.
Treatment assignments to the Latin square are presented in Table 4. This experiment was

conducted in the Fall of 1992.

35

36

Table 4. Final treatmentl assignments for the 4 x 4 Latin square used in experiment
92CSM].

 

Period

 

Cow 1 2 3 4

 

SO L0 S+ L+
L0 L+ SO 8+
8+ SO L+ L0
L+ S+ L0 SO

#mN—t

 

1 Where treatments were coded as: L0 = long cut, 0 WFCS, L+ = long cut, + WFCS, SO
= short cut, 0 WFCS, and 8+ = short cut, + WFCS.

Cow Description

Twelve midlactation cows from the MSU Dairy Farm were reserved for this
experiment. Cows were blocked by parity, days in milk and pre-experiment milk
production. Two squares were composed of multiparous cows and one of primiparous
cows. At the beginning of the experiment, the twelve cows used in the experiment
averaged 125 :l: 32 (mean :1: standard deviation) days in milk, 593 i 60 kg in body weight
and 41.0 :1: 4.6 kg milk produced daily.

Cow Management

Animal use forms were submitted to and approved by the MSU All-University
Committee on Animal Use and Care and the MSU Department of Animal Science.
During the experiment, cows were housed in the activity monitoring unit at the MSU
Dairy Farm. This 12 tie stall setup is located in a barn with a white interior and without
windows. The barn also has continuous lighting and ventilation is driven by exhaust
fans. Cows were milked in place twice daily at 7 am. and 7 pm. Cows were fed once

daily at 9 am. At 9 am, feed tub access doors were closed. After feed tubs were

37
emptied of arts, ﬁlled with fresh TMR, and allowed to stabilize, feed tub access doors
were reopened. This process took approximately 40 minutes. Diets were offered for a
refusal rate at 5 to 10% to insure ad libitum intakes. During nontest weeks, cows were let
out into a cement ﬂoored exercise area by square once daily after one milking for
exercise and a heat check. During test weeks, the cows remained in the stalls. These tie

stalls were bedded twice daily at 5 am. and 8 pm. with shredded phone books.

Forage Harvesting and Preservation

Two alfalfa silages were harvested from a 39 acre ﬁeld of a second cutting mostly
alfalfa forage. The ﬁeld was mowed and conditioned with a sickle bar mower-conditioner
(New Holland Model 1499) and the wilted herbage was chopped the next day with two
forage harvesters (New Holland Model 900): one set for a long cut at a theoretical 3/8"
cut and one set for a short cut at a theoretical 3/16". Alternating sets of four adjacent
windrows, a long cut was harvested ﬁrst then the short cut was harvested. Each load of
herbage from the ﬁeld was weighed in tared wagons before unloading. With one Ag
BaggerTM (Ag Bag Corporation, Astoria, Oregon) ﬁtted with 8' diameter bag, two Ag
BagsTM were ﬁlled and sealed as per manufacturer directions. Stage of maturity was
determined as mean stage by weight (39). Eight samples of 100 to 150 stems were
collected from the ﬁeld and individual stems in each sample were sorted into alfalfa
stage, grass or weed categories. Category samples were dried at 55°C then weighed.
Percentage dry weight for each category within the sample was calculated and then
averaged by category across samples to determine the inputs for the mean stage by weight

formula.

Diet Formulation

Experimental diets were composed of alfalfa silage, grain mix, and WFCS. The
grain mix contained 78.5% dry ground shell corn, 19.6% soybean meal 44, and 1.9%
minerals and vitamins. In diets containing WFCS, 25% of the alfalfa silage NDF was

38
replaced with WFCS NDF. The rations were balanced using Spartan Dairy Ration
Balancer 2 (63) to 27% NDF, 12% Absorbable Protein (using default values for UIP),
and 1% calcium. The diets were formulated to be isoﬁbrous for NDF and isonitrogenous
but not isocaloric. Experimental diets were mixed daily at 7 am. To maintain ration
specs, diet inputs were monitored for changes. Forage samples were taken daily for dry
matter determination. Diet inputs were sampled at the end of each test week and
analyzed for 55°C DM content, ash content, NDF content, and CP content. Diet
reformulation to experimental speciﬁcations occurred as necessary for dry matter and at

least once per period for nutrient analyses.

Data Collected

At each milking throughout the experiment, milk production per cow was
recorded to the nearest .05 kg. Each day at the beginning of the 9 am. feeding, water
meter readings were recorded manually for all cows. At the daily feeding, the as fed
weight of TMR offered and weight of orts from previous day were recorded to nearest .5
kilograms. Body weights to nearest .5 kilograms of all cows were taken after the
morning milking on each of the ﬁrst two days of the period and on the ﬁrst two days
following the end of period four. Consecutive days were averaged to calculate change in
body weight across a period. All cows were given a body condition score (23) to the half
score on a 1 to 5 scale by three independent scorers. Scores were taken on the ﬁrst day of
each period and again at the end of the experiment. The scores at each time point were
averaged across scorers and used to calculate change in body condition score across a

period. Health and reproductive information was recorded at occurrence.

39
Samples Collected

Milk from all cows was sampled for composition analysis at each milking on days
15, 17, and 19 of each period. Morning and evening milk samples were not composited.
After weighing the rrrilk and recording the weight, all milk was poured into another
bucket and then all was poured back again. Using a sampling dipper, milk was collected
and poured into 6 oz. milk sample bag containing a 40 mg potassium dichromate tablet
(Whirl-pakm, Nasco, Fort Atkinson, WI). Samples were sealed and placed in cooler. At
the end of test week, all milk samples were taken to Michigan DHI (East Lansing, M1) for
analysis for fat, protein, lactose and somatic cell count with infrared spectroscopy. Using
DHI milk composition analysis, fat corrected milk (46) and solids corrected milk (Tyrell
and Reid (62), assuming .7% ash) were calculated.

During each test week, diet inputs (.5 kg), experimental diets (.5 kg) and
individual cow orts (one-eighth of daily orts) were collected daily, frozen, and
composited across the test week. At the end of the test week, all composites were dried at
55°C for 48 hours. The entire composite was ground in a Wiley mill (Arthur H. Thomas,
Philadelphia, PA) with a 6 mm screen. The 6 mm grind was subsampled by rolling and
quartering, ground through a Wiley mill with a 1 mm screen (except WFCS where a 2 mm
screen was used for the ﬁnal grind) and stored in a 120 ml screw top plastic specimen cup
at room temperature until analysis. After each grind, the Wiley mills were cleaned and

the residue was collected quantitatively.

Sample Analysis

Diet, diet ingredients and ort samples were analyzed for NDF, ADF and AD
sulfuric acid lignin sequentially (66), CP as N x 6.25 (32), crude fat (ethyl ether
extraction in a modiﬁed Soxhlet apparatus for 24 hours) and ash (ignited at 500°C for 5
h). When possible, duplicates were run on separate days. Duplicates were rejected as

unacceptable when the coefﬁcient of variation was greater than 1% for sequential ﬁbers,

40

2% for CP and 5% for crude fat and ash. Concentrations of all nutrients, except DM,
were expressed as a percentage of DM determined from forced air oven drying at 105°C.
To compare the silages, several additional analyses were performed. On dried ground
samples, insoluble nitrogen (41) and in vitro true digestibility (28) at 0 h, 10 h, 20 h, and
120 h were also run. Wet silage composites were also collected during test weeks and
frozen. These samples were used to quantify silage acids, silage pH, and silage particle
size.

Results of in vitro digestion were used to calculate N DF digestion rate and extent.
The fraction of ﬁber that is indigestible (I) was deﬁned as the residual NDF after 120
hours of in vitro fermentation and was calculated as the 120 hour in vitro NDF residue
divided by zero hour NDF. The fraction of ﬁber that is potentially ferrnentable (D) was
deﬁned as the NDF that disappeared during in vitro fermentation and was calculated as l
- I. The rate of NDF digestion was calculated as:

111 [(X-Z)/D] -1n [(Y-Z)/D)l
rate =

 

10 h.
where:
X = 10 h in vitro fermentation NDF residue
Y = 20 h in vitro fermentation NDF residue
2 = 120 h in vitro fermentation NDF residue

D = the proportion of NDF that disappeared during in vitro fermentation

Water extraction of wet silage samples was used for measurement of silage pH
and silage acid content. A representative wet silage sample of 10.0 grams was blended in
100 ml distilled water with a Sorvall Omni Mixer 17150 (I. Sorvall, Inc., Newtown, CT,
USA) for 45 seconds at speed 10. After blending, pH was measured (Digital Ionizer,
Orion Research, Inc., Cambridge, MA, USA.) and homogenized samples were squeezed

41
through four layers of cheese cloth into two 50 ml centrifuge tubes. Samples were
centrifuged at 20,000x g for 30 minutes. Supernatant was stored frozen until analysis for
acid content. Silage lactic, acetic, propionic and butyric acids were measured on a HPLC.
The column used was an Aminex HPX-87H (Bio-Rad Laboratories, Richmond, CA).
Column temperature was 50°C and solvent was .005 N H2804. Detection was by
refractive index (W aters® 410, Millipore Corp., Milford, MA).

Particle size was determined by wet sieving wet silage containing 5 g of DM
following a macro neutral detergent extraction where the ratio of dry matter to neutral
detergent solution was 5 g DM to 1 L . Samples were wet sieved through the following
screens sequentially: 19000, 9500, 4750, 2360, 1180, 600, 300, 150 and 75 microns.
Percent of dry NDF residues by weight by sieve was used to determine mean particle

size. Variance was determined by ﬁtting results to a gamma distribution (2).

Electronic Data Collection (EDC)

Cow activities were monitored electronically for 5 days beginning on day 15 of
each period. An electronic data acquisition system for continuous monitoring of feed
intake, water intake, and chewing activity of 12 cows (18) was used. The computer
received and processed three signals for each cow: one from a load cell suspending a
feed tub; one from a water ﬂow meter", and one from a pressure sensors on the halter.
The calibrated load cell monitored the changing weight of the hanging feed tub. The
water ﬂow meter monitored the water sent to each cow's drinking cup. The pressure
sensor connected to water ﬁlled bladder, which was attached to the halter under mid jaw,
monitored the pressure changes caused by the movement of the jaw. All signals sent to
the computer were processed and summarized by Labtech NotebookTM to individual cow
period print ﬁles every 5 seconds.

Calibration, adjustment and repair of equipment were done as needed and actions

during test week were recorded for later reference. Daily water meter readings and daily

42
as fed intakes were recorded manually for later comparison to data collected
electronically. Cow trainers in the whole barn were kept off during test weeks as their
pulse of voltage decreased accuracy of water meter reading. Chew halters were placed on
cows one day prior to collection for animal adjustment. Disruption of animal activity was
kept to a minimum during test weeks.

Electronic data collection recorded 48 cow period ﬁles. These raw data were
summarized on an IBM 3090 mainframe computer (lntemational Business Machines
Corporation, Armonk, NY) with an interpretive algorithms in SAS® (see Dado and Allen
(18) for deﬁning parameters). The resulting eating, ruminating and drinking activity
summaries were internally labeled and screened for accuracy using equipment test week
logs and manual intake records. Errant data within a day resulting from faulty equipment
removed the whole cow day from the ﬁnal activity data set. Meals and bouts were
deﬁned and calculated as per Dado and Allen (19). Using postexperiment lab analysis,
dry matter and NDF intakes for each recorded meal was calculated. Corrected (nongrain)
NDF content for each diet was calculated by subtracting grain NDF from the total diet
NDF. Corrected NDF intake was determined by multiplying DMI by corrected NDF
content of diets. With meal information complete, the ﬁnal activity was summarized by
day then averaged by cow and period. Variables summarized included number and
length of eating bouts, total eating chews and time, meal size, number and length of

ruminating bouts, total ruminating chews and time, number and size of drinking bouts.

Statistical Analysis

Cow period daily averages were analyzed as a replicated (n=3) 4 x 4 Latin square
with cows nested within squares using JMP® Version 2 (37; SAS Institute, Inc. Cary,
NC) . Preplanned orthogonal contrasts were used to determine the signiﬁcance of the
main treatment effects of length of cut, WFCS treatment and their interaction. Another

set of preplanned orthogonal contrasts compared squares. Model effects were declared

43
signiﬁcant when P < .05. The degrees of freedom for each source of variation in this
experiment are presented in Table 5. Composition of silage was compared by unpaired t

test on period means. Pearson correlations (37) were run on selected activity variables.

Table 5. Degrees of freedom for each source of variation.

 

 

Source of Variation Degrees of Freedom
Square 2
Orthogonal contrasts:
primiparous vs. multiparous (1)
multiparous 1 vs. multiparous 2 (1)
Cow in square 9
Period 3
Treatment 3
Orthogonal contrasts:
chop length (m, L+ vs. so, S+)1 (1)
whole cottonseed (L+, S+ vs. L0, SO) (1)
interaction (L+, SO vs. LO, 8+) (1)
Error 30
Total 47

 

lWhere treatments were coded as: L0 = long cut, 0 WFCS, L+ = long cut, + WFCS, SO =
short cut, 0 WFCS, and 8+ = short cut, + WFCS.

Calculation of Fiber Effectiveness Coefﬁcient
The calculation of the effectiveness coefﬁcient for WFCS ﬁber was based on the
total chews per kilogram NDF intake of alfalfa silage and WFCS within length of cut.

The formula was:

A-( 1-B)(C)

(B)(C)

where:

44

A = total chews per kilogram corrected NDF intake of diet containing WFCS
B = the measured proportion of alfalfa silage NDF replaced by WFCS NDF.
C = total chews per kilogram corrected NDF intake of diet without WFCS.

The three assumptions in the calculation are: a linear relationship exists between
total chews per day and alfalfa silage NDF intake for the range of intakes in this study, all
chews were due to alfalfa silage NDF and WFCS NDF and not grain mix NDF, and the
number of chews was negligible at 0 NDF intake. For comparison, WFCS effectiveness
coefﬁcients were calculated for each length of cut. Total chews were deﬁned as the sum
of the daily averages of eating and ruminating chews on a given treatment. Kilograms
corrected NDF intake was the total NDF intake minus the NDF associated with the grain
mix. The numerator is the total chews per kilogram corrected NDF intake attributed to
WFCS NDF and the denominator is the total chews per kilogram corrected NDF intake
attributed to alfalfa silage. This calculation is illustrated graphically in Figure 3.

The 95% conﬁdence interval of the true mean for total chews per kilogram
corrected NDF intake was calculated for each treatment. These ranges were used to
calculate the range of eNDF values of WFCS for each length of cut. Also, eNDF
coefﬁcients were calculated in the same manner using total chewing time instead of total

chews.

45

 

 

 

 

E
E
g9 a)
= g Proportion of Alfaﬂa
>4 5 Silage NDF Replaced
:5 by WFCS
LL
9 a
W
3 Z
0 "U
6 8 Proportion of Alfalfa
_, ° Silage NDF Constant
,‘3 g in Both Diets
o o
I— U

 

 

No Added
WFCS WFCS
Diet Diet

Figure 3. Graphical representation of the formula used to partition the total chews per
kilogram corrected NDF intake. The calculation of ﬁber effectiveness is Z divided by Y
where Z = the proportion of chews attributed of WFCS NDF = A—(l-B)(C), Y = the
proportion of chews attributed the replaced alfalfa silage NDF = (B)(C), A = total chews
per kilogram corrected NDF intake of diet containing WFCS, B: the measured
proportion of alfalfa silage NDF replaced by WFCS NDF, and C— - total chews per
kilogram corrected NDF intake of diet without WFCS.

RESULTS

Comparison of Silages

The field used in this experiment was composed of 88.0% alfalfa, 10.2% grass
and 1.8% weeds on a dry matter basis. The alfalfa mean stage by weight was determined
to be 4.6, which is between late bud ("4") and early ﬂower ("5"). About six metric tons
of dry matter were harvested for each length of cut. The long cut silage was twice the
MPS (11.4 mm vs. 5.8 mm, P < .01) of the short cut silage. Measurements made on
samples collected during the experiment (T able 6) revealed the silages were similar (P >
.05) in NDF content (42%), NDF indigestibility (44%), rate of NDF digestion (9%/h),
ADF content (28%), lignin content (6.6%), CP content (22%), insoluble CP content
(9.7%), ash content (9%), and crude fat content (3%). The long cut silage was also wetter
(38% DM vs. 50% DM, P <.01), lower in pH (4.53 vs. 4.85, P <01) and contained more
lactate (6.58% of DM vs. 3.65% of DM, P < .01), more acetate (1.78% of DM vs. .91%
of DM, P < .01) and more propionate (.29 % of DM vs. .20 % of DM, P < .01) than the
short cut silage. Table 7 presents the average percentage across periods of dry weight of
macro NDF residue retained on standard sieves following wet sieving for each silage.
After ﬁtting wet sieving results to a gamma distribution, the standard deviation of particle

size across periods for the long cut was :1: 2.6 mm and for the short cut was i- .7 mm.

46

47

Table 6. Description of long and short cut alfalfa silages.

 

 

 

Length of Cut
long short SE P 1

Kg harvested

wet 32840 23140 - -

dry 12560 11610 - -
55°C DM % 37.6 49.5 .5 **
NDF, % of DM 42.4 41.9 .5 NS
In vitro NDF digestibility, %2 56.3 55.3 1.0 NS
In vitro NDF indigestibility, %3 43.7 44.7 1.0 NS
Rate of NDF digestion(%lhour) 9.5 8.8 .4 NS
ADF, % of DM 28.3 28.4 .5 NS
Lignin, % of DM 6.46 6.70 .22 NS
Mean particle size, mm4 11.4 5.8 .9 **
CP, % of DM 22.6 22.1 .4 NS
Insoluble CP, % of DM5 9.76 9.72 .22 NS
pH 4.53 4.85 .03 **
Acids, % of DM

lactic 6.58 3.65 .16 **

acetic 1.78 .91 .05 **

propionic .29 .20 .02 **

butyric nd6 nd - -
Ash, % of DM 8.4 9.5 .4 NS
Crude fat, % of DM 2.9 2.9 .3 NS

 

1Statistical comparisons were done by unpaired t-test.

2Proportion of original NDF content disappearing after 120 h in vitro fermentation.
3Proportion of NDF residue remaining after 120 h in vitro fermentation.
4As detemrined by wet sieving

SSolvent was borate phosphate buffer as per Krishnamoorthy et al. (41)

6nd = not detected
*P < .05
**P < .01

48

Table 7. Average percentage across periods of NDF residue retained on standard
sieves following wet sieving by length of cut.

 

 

 

Percentage of NDF retained

screen size (um) long length of cut short length of cut
19000 15 4
9500 31 12
4750 31 24
2360 11 32
1180 4 13
600 2 5
300 2 5
150 2 3
75 2 3

 

Ingredient and Diet Composition

Ingredient and diet composition are presented in Tables 8 and 9. Values represent
average ingredient and nutrient composition from across the entire experiment.
Postexperiment analysis of diet composites revealed diet NDF contents higher than
expected. During the experiment, grain mix NDF was estimated to be 10%. However, in
postexperiment NDF analysis, the grain mix NDF content of 15% was observed. The
higher total diet NDF contents are due almost solely to the higher than estimated grain
mix NDF content and, to a lesser extent, higher WFCS NDF content. The
postexperiment analysis of WFCS NDF was 55% compared to the preexperiment value
of 47 %. The reason for this increase is unknown by one may speculate that this is an

error in sampling. All other experiment and postexperiment measurements agree.

49

Table 8. Nutrient composition of forages and concentrates as measured during the
experiment for diet formulation and after experiment on test week composites.

 

 

LONG1 SHORT WFCS Grain mix2

During experiment

55°C DM % 37.2 48.9 89.4 nd3

Ash, % of DM 8.29 8.98 3.60 nd

CP, % of DM 22.96 22.60 20.43 nd

NDF, % of DM 41.07 40.37 47.28 nd
After experiment

55°C DM % 36.7 49.0 89.1 88.6

Crude fat, % of DM 2.95 2.92 19.34 2.88

Ash, % of DM 8.41 9.47 3.84 4.91

CP, % of DM 22.57 22.12 20.06 16.79

NDF, % of DM 41.77 42.23 54.81 14.77

ADF, % of DM 28.28 28.38 35.27 4.15

Lignin, % of DM 6.45 6.70 11.01 .59

 

lLONG = long cut alfalfa silage, SHORT = short cut alfalfa silage, WFCS = whole fuzzy

cottonseed

2Grain mix contained 78.5% dry ground shelled corn, 19.6% soybean meal 44, .5%
dicalcium phosphate, .5% limestone, .5% trace mineral salt, .2% magnesium oxide and
.2% vitamin and trace mineral supplements.

3nd = not determined

50

Table 9. Ingredient and nutrient composition of diets by treatment as balanced during
experiment and as measured postexperiment on test week composites.

 

 

 

Treatmentl
L0 L+ 80 8+ SE

Ingredients, % DM of diet

Alfalfa hay crop silage 55.29 40.99 55.29 40.99

WFCS - 11.89 - 11.89

Grain Mix2 44.72 46.69 44.72 46.69

Limestone — .43 - .43
Composition, as balanced during experiment

55°C DM % 49.20 55.63 59.60 65.15

CP, % of DM 20.17 19.76 20.17 19.76

Absorbable CP, % of DM 11.49 12.06 11.49 12.06

NDF, % of DM 26.98 26.92 26.98 26.92

Ca, % of DM 1.03 1.00 1.03 1.00
Composition, as measured postexperiment

55°C DM % 50.1 56.3 60.9 66.1 1.11

Crude fat, % of DM 3.26 5.29 2.93 4.90 .26

Ash, % of DM 6.87 6.68 7.62 7.05 .40

CP, % of DM 20.01 19.25 19.44 19.63 .65

NDF, % of DM 29.64 30.95 30.90 30.75 .91

ADF, % of DM 17.30 18.15 18.05 17.88 .65

Lignin, % of DM 3.70 4.20 3.99 4.32 .25

Corrected NDF, % of DM3 23.03 24.05 24.29 23.85 nd4

 

lTreatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, SO = short

cut, no WFCS, and 8+ = short cut, added WFCS.

2Grain mix contained 78.5% dry ground shelled corn, 19.6% soybean meal 44, .5%
dicalcium phosphate, .5% limestone, .5% trace mineral salt, .2% magnesium oxide and

.2% vitamin and trace mineral supplements.
3corrected NDF = Total NDF - Grain NDF

4nd = not determined.

51
Milk Production and Composition
Milk production and composition data are presented in Table 10. Length of cut
did not effect (P < .05) milk yield or composition. Cottonseed addition increased milk
yields (35.5 kg vs. 33.7 kg, P < .01) but did not effect (P > .05) milk composition.
Cottonseed addition also increased (P < .05) 3.5% FCM (35.4 kg vs. 33.5 kg), 4.0% FCM
(32.8 kg vs. 31.1 kg), and SCM (33.1 kg vs. 31.5 kg). There were no interactions (P >

.05) between length of cut and WFCS treatment on milk yield or composition.

Feed Intake, Body Weight and Condition, and Cow Health

Intake and body weight and condition data are presented in Table 11. Manual (7
day averages) DM and water intakes agree closely with electronic data collection (EDC)
intakes (5 day averages). Long length of cut had lower EDC DMI (23.5 kg vs. 24.3 kg, P
< .05) and lower water intakes (85 L vs. 92 L, P < .01) when compared to short length of
cut. Added WFCS increased EDC dry matter (24.8 kg vs. 23.0 kg, P < .01) and water
intakes (90 L vs. 86 L, P < .01). NDF intake was proportional to DMI. An interaction of
length of cut and WFCS (P < .05) made the difference of EDC DMI of the two short cut
diets (2.7 kg) greater than the difference of the two long cut diets (.8 kg). Change in body
weight and change in body condition score across periods were unaffected (P > .05) by
length of cut or WFCS treatment. Incidence of health problems is presented in Table 12.
With the exception of one displaced abomasum, all cows were healthy during the
experiment. The cow with the displaced abomasum was not removed from the study, but

only one day of data was used for the period in which the displaced abomasum occurred.

52

Table 10. Effect of treatment on milk production and composition means.

 

Treatmentl Effect2

 

 

L0 L+ SO S+ SE L C L X C

 

Milk Yield, kg/d

Actual 34.3 35.7 33.0 35.2 6 NS ** NS
FCM, 3.5% 34.1 35.6 32.9 35.1 8 NS * NS
FCM, 4.0% 31.7 33.0 30.5 32.6 7 NS * NS
SCM 32.0 33.2 30.5 33.0 7 NS * NS
Milk Composition
Fat, % 3.48 3.49 3.54 3.49 .08 NS NS NS
Protein, % 3.27 3.27 3.21 3.30 .05 NS NS NS
Lactose, % 4.85 4.78 4.77 4.88 .05 NS NS NS
SNF, % 8.84 8.76 8.69 8.89 .09 NS NS NS
SCC, 1000's/ml 62 61 56 60 7 NS NS NS

 

lTreatment codes are: L0 = long cut, no WFCS, L-t- = long cut, added WFCS, SO = short
cut, no WFCS, and 8+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L X C =
Interaction of length of cut and WFCS treatment.

*P < .05
**P < .01.

53
Table 11. Effect of treatment on intake, body weight, and body condition.

 

Treatmentl Effect2

 

 

L0 L+ SO S+ SE L C L X C

 

Dry Matter Intake, kg/d
Manual 23.3 25.0 23.0 25.8 .7 NS ** NS

EDC3 23.1 23.9 22.9 25.6 .4 * ** *
Water Intake, L/d

Manual 82.4 87.5 89.6 94.8 1.2 ** ** NS

EDC 82.0 86.9 89.6 94.1 1.3 ** ** NS
NDF Intake, kg/d

EDC 6.8 7.4 7.1 7.9 .1 ** ** NS

corrected“ 5.3 5.7 5.5 6.1 .1 * ** NS
Body Wt. Change, kg 11 13 10 17 5 NS NS Ns
Body Condition Change5 .11 .08 .02 .08 .06 NS NS NS

 

lTreatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, SO = short
cut, no WFCS, and 8+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L X C =
Interaction of length of cut and WFCS treatment.

3EDC = electronic data collection
4corrected NDF intake = total grain NDF minus grain NDF
5on 1 to 5 scale (23)

*P < .05
**P < .01.

54

Table 12. Summary of animal health data during experiment listing the number of
occurrences during each treatment.

 

Treatmentl

 

Problem L0 L+ SO S+

 

Subdennal milk vein rupture
Traumatized teat end
Displaced abomasum

Cystic ovaries

Clinical mastitis

Clinical ketosis

OO—‘OI—O
OOOOOO
OOOr-‘OO
OOOOOt—d

 

lTreatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, S0 = short
cut, no WFCS, and 8+ = short cut, added WFCS.

55

Chewing Activity

Eating, ruminating and total activity means by treatment are presented in Tables
13, 14 and 15, respectively. All 48 cow periods are represented and 92% (220 out of
240) of the cow days were successfully collected. Eating activity was affected by length
of cut and WFCS treatment. Long length of cut increased eating chews when measured
per day (17197 vs. 15520, P < .01), per eating bout (1798 vs. 1534, P < .05), per
kilogram DMI (742 vs. 650, P < .01), per kilogram NDF intake (2460 vs. 2108, P < .01),
and per kilogram corrected NDF intake (3169 vs. 2711, P < .01). Long length of cut
increased eating time in minutes per day (273 vs. 248, P < .01), per eating bout (28.4 vs.
24.6, P < .01), per kilogram DMI (11.8 vs. 10.4, P < .01), per kilogram NDF intake (39.0
vs. 33.7, P < .01), and per kilogram corrected NDF intake (50.2 min vs. 43.4 min, P <
.01). Length of cut did not effect (P > .05) number of eating bouts per day (10.5), meal
size in kilogram dry matter (2.4 kg) or kilogram NDF (.73 kg) or interrneal interval (115
min). The addition of WFCS to treatment diets decreased eating chews as measured per
day (15654 vs. 17064, P < .01), per kilogram DMI (638 vs. 752, P < .01), per kilogram
NDF intake (2070 vs. 2498, P < .01), and kilogram corrected NDF intake (2671 vs. 3209,
P < .01). The addition of WFCS to treatment diets also decreased eating time in minutes
per day (252 min vs. 270 min, P < .01), per kilogram DMI (10.2 min vs. 11.9 min, P <
.01), per kilogram NDF intake (33.3 min vs. 39.4 min, P < .01), and per kilogram
corrected NDF intake (43.0 min vs. 50.6 min, P < .01). However, the addition of WFCS
did not effect (P > .05) the number of eating chews per bout (1666), minutes per eating
bout (26.5 min) or number of eating bouts per day (10.5). The addition of WFCS
increased average meal size in kilogram dry matter (2.5 kg vs. 2.4 kg, P < .05) and
kilogram NDF (.76 kg vs. .70 kg, P < .01). Cottonseed treatment had no effect (P > .05)
on eating chew rate (61.6 chews per min) or interrneal interval (115 min). A length of cut
and WFCS interaction was observed for daily measures of eating chews (P < .05) and

time (P < .01). Eating chews and time on short diets (15678 chews and 250 min for short

56

cut, no WFCS diet vs. 15363 chews and 247 min for short cut, added WFCS diet) were
close, whereas, eating chews and time on long cut diets were not (18449 chews and 289
min for long cut, no WFCS diet vs. 15945 chews and 257 min for long cut, added WFCS
diet). However, dividing by unit of intake removed the interaction.

Ruminating activity as affected by treatment is presented in Table 14. Long
length of cut had more ruminating chews and time per day than short length of cut (30484
vs. 28062; 491 min vs. 460 min, respectively, P < .01). Relationships continued when
corrections were made for DMI, NDF intake and corrected NDF intake. Ruminating time
per bout and chews per bout were similar across treatments (32.2 min and 1978 chews,
respectively, P > .05) but long length of cut had one more ruminating bout per day (15.4
vs. 14.6, P < .05). The increase in ruminating bouts per day reduced the mean time
between ruminating bouts (61.6 min vs. 68.0 min, P <.01). The addition of WFCS to the
diets did not change (P > .05) ruminating chews per day (29273), ruminating chews per
kilogram DMI (1239), ruminating time per day (475.3 min), ruminating chews per bout
(1978), ruminating time per bout (32.2 min), ruminating bouts per day (15.0), or the mean
time between ruminating bouts (65 min). However, the addition of WFCS reduced
ruminating chews per kilogram NDF intake (3927 vs. 4204, P < .05), ruminating chews
per kilogram corrected NDF intake (5068 vs. 5407), ruminating time per kilogram DMI
(19.6 min vs. 20.7 min, P < .01), ruminating time per kilogram NDF intake (64.6 min vs.
68.8 min, P < .01), and ruminating time per kilogram corrected NDF intake (82.0 min vs.
88.5 min, P < .01). An interaction of length of cut and WFCS treatment was observed for
ruminating chews per day (P < .01), ruminating time per day (P < .05), and ruminating
chews per kilogram NDF intake (P < .05). Ruminating chew rate was unaffected by
treatment (61.01 chews/min, P > .05).

Total chewing activities, the sum of eating and ruminating activities, are reported
in Table 15. Long length of cut had greater total chews and total chewing time (P < .01)
per day (47680 vs. 43582 and 764 min vs. 708 min, respectively), per kilogram DMI

57
(2046 vs. 1823 and 32.8 min vs. 29.6 min, respectively), per kilogram NDF intake (6780
vs. 5920 and 108.6 min vs. 96.3 min, respectively), and per kilogram corrected NDF
intake (8740 vs. 7614; 140.2 min vs. 128.8 min, respectively). Addition of WFCS did not
change total chews or total chewing time (45631 chews, 736.0 min, respectively, P >
.05), but did reduce total chews per kilogram DMI ( 1850 vs. 2020, P < .01), total chews
per kilogram NDF intake ( 5998 vs. 6702, P < .01), total chews per kilogram corrected
NDF intake ( 7738 vs. 8616, P < .01), total chewing time per kilogram DMI (29.8 min vs.
32.6 min, P < .01), total chewing time per kilogram NDF intake (96.8 min vs. 108.2 min,
P < .01), and total chewing time per kilogram correcting NDF intake (125.0 min vs.
139.0 min, P < .01).

An interaction was observed for length of cut and WFCS treatment for total chews
and total chewing time per day (P < .01). For the long length of cut, the addition of
WFCS decreased total chews per day (49376 vs. 45985) and total chewing time per day
(785 min vs. 742 min) and, for short length of cut, the addition of WFCS increased total
chews per day (42492 vs. 44671) and total chewing time per day (697 min vs. 719 min).
The interaction was not observed when total chews and total chewing time were divided
by kilograms DMI. Length of cut and WFCS treatment interaction was observed (P <
.05) for total chews per kilogram NDF intake and total chews per kilogram corrected
NDF intake with the effect of cottonseed addition being greater for the long cut than the
short cut. A trend (P < .10) for a similar interaction was found for total chewing time per

kilogram NDF intake and per kilogram corrected NDF intake.

58

Table 13. Eating activity as affected by treatment.

 

Treatment1 Effect2

 

 

L0 L+ SO S+ SE L C L X C

 

Eating Chews
M 18449 15945 15678 15363 527 ** * *
lbout 1940 1657 1563 1504 102 * NS NS
lkg DMI 810 673 695 604 27 ** ** NS
/kg NDFI 2744 2177 2252 1964 87 ** ** NS

/kg corrected NDFI 3527 2811 2891 2531 113 ** ** NS

Eating Time, min

/d 289 257 250 247 5 ** ** **
lbout 30.1 26.7 24.9 24.2 1.2 ** NS NS
lkg DMI 12.7 10.8 11.1 9.7 .3 ** ** NS
/kg NDFI 42.9 35.0 35.8 31.6 1.0 ** ** NS
/kg corrected NDFI 55.1 45.2 46.0 40.7 1.3 ** ** NS
Eating Bouts, /d 10.4 10.2 10.6 10.6 .3 NS NS NS
Meal Size
kg DM 2.4 2.5 2.3 2.5 .1 NS * NS
kg NDF .70 .77 .70 .76 .02 NS ** NS
Chew Rate, chews/min 62.6 61.0 61.3 61.3 1.3 NS NS NS
Intermeal Interval, min 114.7 118.2 113.8 112.9 4.3 NS NS NS

 

lTreatment codes are: L0 = long cut, no WFCS, L4» = long cut, added WFCS, SO = short
cut, no WFCS, and 8+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L X C =
Interaction of length of cut and WFCS treatment.

*P < .05
**P < .01.

59

Table 14. Ruminating activity as affected by treatment.

 

Treatmentl Effect2

 

 

L0 L+ SO S+ SE L C L X C

 

Ruminating Chews

Id 30927 30040 26815 29308 612 ** NS **
lbout 1996 1987 1914 201 1 48 NS NS NS
/kg DMI 1347 1263 1187 1160 30 ** NS NS
lkg NDFI 4563 4077 3846 3777 103 ** * *

lkg corrected NDFI 5876 5266 4938 4869 136 ** * NS

Ruminating Time, min

/d 496 486 447 472 7 ** NS *
/b01.lt 32.1 32.2 31.9 32.4 .7 NS NS NS
Ikg DMI 21.6 20.5 19.8 18.7 .4 ** ** NS
/kg NDFI 73.4 66.2 64.1 60.9 1.4 ** ** NS
/kg corrected NDFI 94.6 85.5 82.4 78.6 1.9 ** ** NS
Ruminating Bouts, /d 15.6 15.2 14.4 14.8 .3 * NS NS
Chew Rate, ChCWS/min 61.8 61.3 59.6 61.4 .8 NS NS NS
Interrurrrinating Interval, min3 60.5 62.8 70.1 65.8 1.8 ** NS NS

 

lTreatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, SO = short
cut, no WFCS, and 8+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L x C =
Interaction of length of cut and WFCS treatment.

3Interruminating interval is the elapsed time between the end of one ruminating bout and
the start of next ruminating bout.

*P < .05
**P < .01.

60
Table 15. Total activity as affected by treatment.

 

 

 

 

Treatmentl Effeth
L0 L+ SO S+ SE L C L X C
Total Chews
M 49376 45985 42492 44671 902 ** NS **
/kg DMI 2157 1936 1882 1764 47 ** ** NS
/kg NDFI 7307 6254 6098 5741 163 ** ** *
/kg corrected NDF] 9404 8077 7829 7400 216 ** ** *
Total Chewing Time, min
/d 785 742 697 719 7 ** NS **
/kg DMI 34.3 31.3 30.8 28.4 .6 ** ** NS
lkg NDFI 116.2 101.1 100.0 92.5 2.1 ** ** NS
Ikg corrected NDFI 149.7 130.6 128.4 119.3 2.8 ** ** NS

 

1Treatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, SO = short

cut, no WFCS, and 8+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L X C =

Interaction of length of cut and WFCS treatment.
*P < .05
**P < .01.

61

Drinking Activity

Drinking activity as affected by treatment is presented in Table 16. All 48 cow
periods are represented and 98% (236 out 240) of the cow days represented. Short length
of cut increased water intake (91.8 L/d vs. 84.4 L/d, P < .01), increased the number of
drinking bouts per day (16.6 vs. 15.3, P < .01), and time spent drinking each day (17.2
min vs. 15.7 min, P < .01). Short length of cut diets also had a shorter interval between
drinking bouts (94 min vs. 104 min, P < .05). WFCS addition increased water intake
(90.5 L vs. 85.8 L, P < .01), increased drinking time (17.1 min vs. 15.8 min , P < .01),
and increased drinking bout length (1.18 min vs. 1.08 min, P < .05). Interaction between
length of cut and WFCS treatment was observed for daily drinking time (P < .05).
Length of cut and WFCS treatment had no effect (P > .05) on drinking rate (5.35 L/min.)
or drinking bout size (6.17 L).

Correlations

Pearson correlation coefﬁcients among the 48 cow period means of 22 feeding
activity variables are presented in Table 17. Due to the experimental design, a high
correlation was found between DMI and NDF intake (r = .96) and, therefore, the
variables produced when any feeding activity was divided by both DMI and NDF intake
had a similar correlation coefﬁcients with any other variable of comparison. Measures of
number of chews per day and time spent chewing within a type of activity were highly
correlated (r = 2.85). The three ways used to measure eating chews (per day, per
kilogram DM and per kilogram NDF) were highly correlated (r = 2.85) with each other.
All activities (eating, ruminating and total) measured with number of chews and time
spent chewing were highly correlated (r = 2.85) with each other when divided by
kilogram DM and kilogram NDF.

62

Table 16. Drinking activity as affected by treatment.

 

 

 

 

Treatmentl Effect2

L0 L+ SO S+ SE L C L x C
Water Intake, L/d 82.0 86.9 89.6 94.1 1.3 ** ** NS
Drinking Time, min/d 15.4 16.0 16.2 18.2 .3 ** ** *
Drinking Bouts, /d 15.8 14.8 16.6 16.6 .4 ** NS NS
Drinking Bout Size, L 5.9 6.5 6.1 6.1 .2 NS NS NS
Drinking Bout Length, min 1.08 1.19 1.09 1.1 .04 NS * NS
Drinking Rate, L/min 5.3 5.4 5 5 5.2 .1 NS NS NS
Interdrink Interval, min 102.9 104.5 97 2 90.9 4.2 * NS NS

 

1Treatment codes are: L0 = long cut, no WFCS, L+ = long cut, added WFCS, 80 = short

cut, no WFCS, and S+ = short cut, added WFCS.

2L = Main effect of length of cut; C = Main effect of WFCS treatment; L x C =

Interaction of length of cut and WFCS treatment.
*P < .05
**P < .01.

63

.ri 55 32 20 mm.
:2: 53on 352-8.» .28 Amovnc 522028 2:3ng .50. 55 m3— 8 mm. :2: 83on $32.23 8m COVE cease-Bo 2505382

 

 

8. 3. ea. 8. em. 3. ca. 2. 8. mm. 3.. mm. on- em. 2. 2.. em. 32. en. E.- e? ”52 SEE. as
8. 8. es. mm. 8. x... mm. 8. as. 3.. mm. 2.. 2. E. 2.. ex. 2. en. em- 2: 2o 9:5... :2
E. E. a. S. E. em. 2. we. em. em. a: mm. mm. me. 8. 8. E. 8.- 8.- 25:. so
2:: worsens :28.
ms. 2. 3. ex. R. a. 8. em. a. 2.- em. 3. E. 8. 3. E. $.- ee. "52 3 a:
E. em. 3. on. a. a. a. a. 2.- a. E. mm. S. 2. E. 2..- 3.. 2o i a:
2. E. R. me. E. ow. mm. 8.- 2.. E. B. B. S. we. 8.- S.- Q E:
950:0 39H
8. we. 8. 2. cm. 8. 2.- me. me. a. we. 8. 8. en- E.- ..Ez ESE: 6:
2. a. 8. mm. em. 2.- a. G. 2. 3. me. E. K: me. 2o 335 a:
S. E. 2. on. 8.- 2. 2. 2. mm. a. a. 8.- 8.- Res 3.:
0E: wﬁﬁﬁEzm
Ms. :. em. 8.. ee. 3. mm. 2. so. 2.. 8.. _e.- "52 me 222
E. em. 2.- am. 8. a. me. E. 2.. on- 8.- 2o 3 a:
R. 2.- 2. 2. 2. mm. on. 2.. we. 8. Q 2:
255 2.825.55—
2.- 2. ma. em. 2. 2. 2. 2.- :.- 222253. a:
:..- 9..- 3..- mm.- 2.- 2.- 2. 2. 2:2 2
2525
we. R. 3. 8. ms. so; 2.- "52 BEE Q
R. 8. a. E. B.- ee: So 322. 6
me. E. 8. 8.- 8.- Res 2
as. 3.2
as. em. $.- 2: "52 3 6
em. we- we- :5 me E
2.- 2.- Q 6
@326 weuam
em. 2232 6
2.3.35 8
oase—
22 so a: a: EU a: a: a: a: a: c: 8: 2 Q E 62 5 as 6 Q E

 

238%

_.mo_nmtd> .0328 mam—ouch 252-? .«o 282: eaten Boo we macaw 256508 592028 com-Eon .2 03m...

64
Results of eNDF Calculations

Results of ﬁber effectiveness calculations are shown in Table 18. The measured
proportion of alfalfa silage NDF replaced by WFCS NDF was .27. Using total chews per
kilogram corrected NDF intake, effectiveness coefﬁcients were calculated as .48 for
WFCS NDF in long alfalfa silage and .80 for WFCS NDF in short alfalfa. To calculate
the range for eNDF coefﬁcients, the 95% conﬁdence interval of the true treatment means
were determined as the measured treatment means 2137 chews. Using these ranges to
calculate the ranges of effectiveness coefﬁcients, the eNDF of WFCS in long cut alfalfa
silage was maximum at .58 and minimum at .38, and the eNDF of WFCS in short cut
alfalfa silage was maximum at .93 and minimum at .67. The effectiveness of short alfalfa
NDF relative to long alfalfa NDF was calculated as .835 :t: .025 (7829 (= total chews per
kilogram corrected NDF intake for short cut, no WFCS) + 9404 (= total chews per
kilogram corrected NDF intake for long cut, no WFCS».

Using similar equations, total chewing time per kilogram corrected NDF intake
was used to calculate effectiveness coefﬁcients. eNDF of WFCS in long alfalfa silage
was .53 :1: .08, eNDF of WFCS in short alfalfa silage was .74 i .10 and short alfalfa silage
NDF relative to long alfalfa silage NDF was .86 i .02.

Table 18. Fiber effectiveness coefficients calculated in this experiment.

 

Calculated by Total Activity as recorded

 

 

by Chews by Time
Fiber effectiveness of WFCS
in long length of cut .48 .53
Fiber effectiveness of WFCS
in short length of cut .80 .74

Fiber effectiveness of short length of cut relative
to long length of cut .84 .86

 

65
Comparison of Responses by Square

A comparison of responses by square is presented in Table 19. Preplanned
orthogonal contrasts revealed differences between the primiparous square (square 3) and
the multiparous squares (square 1 and 2) and between the two multiparous squares.

When compared with the multiparous squares, the primiparous square produced
less milk (32.7 kg vs. 35.5 kg, P < .01) of similar milk fat content (3.5%, P > .05), ate
less DM (23.0 kg vs. 24.4 kg, P < .01), less NDF (7.0 kg vs. 7.4 kg, P < .01), and less
corrected NDF (5.4 kg vs. 5.8 kg, P < .01) per day in larger meals (2.6 kg DM vs. 2.3 kg,
P < .01), and drank less water (80.4 L/d vs. 92.0 L/d, P < .01) in less time (15.6 min/d vs.
16.9 min/d, P < .01). The primiparous square ate and drank fewer times per day (9.4 vs.
11.0 and 14.8 vs. 16.6, respectively, P < .01) but ruminated more times per day (16.3 vs.
14.4, P < .01). The primiparous square chewed more times per day and per kilogram
corrected NDF intake (P < .01) while eating, ruminating and overall.

A comparison of the multiparous squares revealed that Square 2 produced more
milk (37 kg vs. 34 kg, P < .01) of similar milk fat (3.5%, P > .05), ate (P < .05) and drank
(P < .01) more total in fewer (P < .01) bouts per day than Square 1. Chewing activity per
day was similar (P > .05) between these two squares, but the higher intake of the Square
2 reduced (P < .05) ruminating and total chews per kilogram corrected NDF].

Because treatment effects were signiﬁcant (P < .05) within square 2 and 3 , eNDF
coefﬁcients were calculated within square. The eNDF coefﬁcient of WFCS NDF in long
cut alfalfa silage were .60, .49, and .40 for Squares l, 2, and 3, respectively. The eNDF
coefﬁcient of WFCS NDF in short cut alfalfa silage were .92, .86, and .64 for Squares l,
2, and 3, respectively. The eNDF coefﬁcient of short cut alfalfa silage relative to long cut

alfalfa silage NDF were .83, .87, and .81 for Squares 1, 2, and 3, respectively.

66

Table 19. Comparison of responses by square.

 

 

 

 

Square Effect1
1 2 3 SE M vs. P 1 vs. 2

Parity2 M M p - - -
DIM3 99 148 129 - - -
Body Weight, kg4 628 598 551 - - -
Milk Yield. mm 34.0 36.9 32.7 .5 ** **
Milk Fat, % 3.5 3.5 3.5 .1 NS NS
Dry Matter Intake. kg/d 23.8 24.9 23.0 .3 ** *
NDF intake, kg/d 7.3 7 .6 7.0 .1 ** *
Corrected NDF intake, kg/d 5.6 5.9 5.4 .1 ** :1:
Meal Size, kg DM 2.1 2.5 2.6 ,1 ** 21:21:
Water Intake. L/d 87.5 96.5 80.4 1.1 ** **
Drinking Time- min/d 18.4 15.3 15.6 .3 ** **
Bouts/d

eating 11.9 10.2 9.4 .3 ** **

ruminating 14.7 14.0 16.3 .3 ** NS

drinking 18.9 14.3 14.8 .3 ** **
Chews/d

eating 15431 15483 18162 457 ** NS

ruminating 291 18 27968 30731 530 ** NS

total 44548 43452 48893 781 ** NS
Chews/kg corrected NDF intake

eating 2764 2639 3418 98 ** NS

ruminating 5207 4793 571 1 l 18 ** *

total 7971 7432 9129 187 ** *

 

1Orthogonal contrasts comparing squares.

2M = multiparous; P = primiparous.

3Average days in milk at start of experiment

4Average body weight at start of experiment

*P < .05
**P < .01.

DISCUSSION

Ingredients and Diets

The particle sizes of the two silages were different (P < .05). Theoretical cut
matched measured particle size reasonably well (9.5 mm and 4.8 mm vs. 11.4 mm and
5.8 mm, respectively). The long cut was attained with 3/8" theoretical cut and had more
long particles (>9500 microns) by weight than the short cut at 3/ 16" (Table 7). While
different in particle size, these two silages are not a typical of silages found on Michigan
dairy farms. The theoretical cut of 3/8" is the recommended length of cut for alfalfa
silage as it provides enough long particles for chewing (52) and is still packable (49).

The material resulting from this theoretical cut is representative of a silage that is reduced
little in particle size by the storage and handling between ﬁeld chopping and cow feeding,
such as with a bunker silage. The short length of cut at 3/ 16" could be considered typical
of 3/8" material that was reduced in particle size between harvesting and feeding as with
an upright bottom unloading silo that cuts material out,.

Because forages were chopped sequentially and not simultaneously, the long cut,
harvested ﬁrst, was wetter than the short cut, harvested second. Still, the DM contents
were appropriate for material put up in trench silos (long cut) and upright silos (short cut).
The two silages of this experiment differed in measured pH and acid content. These
differences may be due to the difference in dry matter content of the silages as drier
conditions inhibit bacteria or to the possible difference of sugar content at ensiling due to
time of respiration in the ﬁeld differences. However, both silages were of good quality

and were well preserved as judged by smell and pH at the given dry matters (49). These,

67

68
along with the similarities in NDF content and NDF digestibilities give support to the
assertion that, with regard to chewing activity, the difference between the silages was
particle size.

Theoretical length of cut was set on the chopper, but several factors in addition to
theoretical cut determine the final particle size of the chopped herbage. They include
sharpness of knives, speed (rpm) of machinery operation, and rate (weight/min) of
herbage pick up (52). Since these cannot be standardized across studies, the
measurement of MPL is appropriate (52). Contrary to most reported studies, particle size
in this study was determined with wet sieving. Most studies use dry sieving methods,
therefore, it is difﬁcult to compare forage particle sizes between this and other studies.

Type of forage, method of processing, and sieving method inﬂuences the ﬁnal
measurement of particle size (24). Dry sieving gives lower (P < .01) mean particle sizes
than wet sieving (2, 24). An explanation for this difference can be found in how the two
methods sort material. Wet sieving, the slurrying of wet suspensions through
standardized sieves arranged in decreasing pore size top to bottom, tends to sort material
by length (65) and, in contrast, dry sieving, done by sifting material via mechanical
shaking of wet or dry material through the standard sieves, tends to sort according to
cross-sectional diameter (65). Gamma distributions ﬁt feedstuff particle distributions
better than log normal (2) and were used to determine the variances.

This study departed from the convention of measuring MPL with dry sieving
because it was felt that particle distributions obtained by wet sieving may be more
meaningful relative to physical processes occurring in the rumen (2). Physical form of
the diet is a major factor stimulating rumination (74). Welch and Hooper (70) suggest
that rumination is stimulated primarily by ingesta particles 2 10 mm. They also note that
particles between 1 and 10 mm in length affect rumination, but are not as important as

longer material in stimulating rumination. Therefore, wet sieving may be more

69

appropriate as it sorts by length. Wet sieving is more tedious and less mechanized than
dry (24) explaining why wet sieving is less used even though it may be more meaningful.

In this experiment, wet sieving was done following a macro NDF extraction. Due
to surface area effects, proportionally, more soluble material would be removed from the
smaller particles during the ﬁve to ten minutes of the wet sieving process than from the
larger particles. Pretreatment by macro NDF eliminated this problem by uniformly
extracting detergent soluble materials from all particles.

The protocol of this experiment was to feed 27% NDF diets with the addition of
WFCS NDF replacing 25% of the alfalfa silage NDF. The NDF of treatment diets of
27% was selected to allow possible differences that would be masked by overfeeding
NDF. Experimental diets were measured at 30% NDF. This discrepancy was the result
of the grain mix NDF content being 5% higher than estimated. The measured
replacement of alfalfa silage NDF with WFCS NDF was 27%. This 2% increase above
planned was due to the increase of WFCS NDF content from 47% to 55%. Measured CP
content of experimental diets was as balanced. Final calcium content of the experimental
diets was not measured. All other measured components were similar across

experimental diets.

Cow Performance

The addition of WFCS to experimental diets increased milk production (actual,
FCM and SCM). This is consistent with the increased energy available to the animals
due to increased DMI and to increased energy density of the WFCS added diets. Short
length of cut and addition of WFCS increased water intake. These diet ingredients made
the experimental diets drier which may have been responsible for the increased daily
water consumption. Also, since milk is 87% water, greater production with added WFCS
diets may also have drove increased daily water consumption. Increased drinking time

was recorded on diets containing the short length of cut and added WFCS. Short length

70
of cut increased drinking time by increasing the number of drinking bouts per day. The
addition of WFCS increased daily drinking time by lengthening of each drinking bout.
Across diets, the cows on experiment gained body weight and body condition

suggesting they were in positive energy balance. Regardless of diet, the cows gained 51
kg of body weight and .29 condition points during the experiment. Animals were healthy
throughout the experiment. Only one health related problem (see Table 12) was known
to effect measurements taken, a displaced abomasum diagnosed on Day 20 of short cut,
no WFCS diet period. Data were dropped from Day 16, the day before animal behavior

was noted as abnormal.

Chewing Activity

An increase in chewing suggests an increased dietary "scratch factor" resulting
from increased ﬁber effectiveness. Increased chewing activity leads to greater saliva
ﬂow to the rumen. In this experiment, the long length of cut alfalfa silage had more
chewing activity than short length of cut. The long cut was ingested at slower rate
requiring more time and chews per unit of dry matter to process the diet to swallowable
boluses. While all diets had a similar number of meals per day (10.4), the meals of the
long cut diets were greater in minutes and in number of chews resulting in greater total
daily eating time and chews. The long cut silage added one more 30 minute ruminating
bout to each day. The addition of one more ruminating bout increased daily ruminating
time and chews, decreased interruminating intervals, and increased ruminating activity
per unit of intake. The increased eating and ruminating activities on the long length of
cut of alfalfa silage summed to greater total chewing activity per day.

The addition of WFCS NDF to the experimental diets to replace alfalfa silage
NDF lowered chewing activity. The addition of WFCS did not change the number of
eating or ruminating bouts per day. While ruminating activity per bout was similar with

and without the addition of WFCS, the addition of WFCS made eating bouts shorter with

71
fewer chews. The addition of WFCS to the experimental diets reduced eating and
ruminating activity per kilogram intake (DM, NDF, and corrected NDF) and,
subsequently, reduced total chewing activity per kilogram intake measured in minutes
and chews. The addition of WFCS increased meal size (intake per eating bout) as
measured in kilogram DM or kilogram NDF. Rate of chewing for eating and ruminating
was constant across the experiment diets at 61 chews/min. The interactions of length of
cut and WFCS treatment caused the difference in total chewing activity divided by intake
between long cut, no WFCS and long cut, added WFCS diets to be greater than the
difference between short cut, no WFCS and short cut, added WFCS diets.

Chewing activities were correlated. Since chew rates were generally 61 chews
per min, chews per day, per kilogram DMI and per kilogram NDFI were highly correlated
to minutes per day, per kilogram DMI and per kilogram NDFI, respectively. Total chews
per kilogram NDF and total chewing time per kilogram NDF were highly correlated (r =
.96) suggesting that the use of chews is equivalent to time as an index for the

measurement of effectiveness when both are determined with this EDC system (18).

Comparison of Responses by Square

Multiparous Square 2 consumed more dry matter and water and produced more
milk. Primiparous Square 3 consumed less dry matter and water and produced less milk.
The multiparous squares processed more material with fewer chews than the primiparous
square suggesting greater efﬁciency. To a lesser extent, Multiparous Square 2 was also
more efﬁcient than Multiparous Square 1. A comparison of eNDF coefﬁcients by square
showed that WFCS NDF was least effective in Primiparous Square 3 and more effective
in Multiparous Square 1. The differences in effectiveness coefﬁcients may be due to the
differences in body size or in chewing efﬁciency by square.

A comparison of this experiment's multiparous and primiparous cows to those of

Dado and Allen (19) show some similarities and differences. Dado and Allen fed a corn

72
silage and alfalfa silage based diet, while the cows on this experiment received an all
alfalfa silage forage base. Their cows were earlier in lactation (63 DIM vs. 125 DIM) and
their multiparous cows consumed a similar amount of dry matter (24.8 kg vs. 24.4 kg) in
an similar number of meals (11 per day). However, they produced more milk (37.5 kg
vs. 35.5 kg) per day than the cows on this experiment. Dado and Allen's multiparous
cows had more eating chews (19256 vs. 15457) and similar ruminating chews (28946 vs.
28543) for a greater total chews per day (48201 vs. 44000). Dado and Allen's
primiparous cows consumed less dry matter (20.0 kg vs. 23.0 kg) and produced less milk
(28.7 kg vs. 32.7 kg) on more meals per day (11.3 vs. 9.4) than the cows on this study.
Dado and Allen's primiparous chewed per day equally while eating (18276 vs. 18162)
and less while ruminating (29645 vs. 30731) for lower total daily chews (47921 vs.
48893). Overall, the activity measures of the cows on this experiment fell between the

activities of the primiparous and multiparous cows of Dado and Allen (19).

Comparison of Chewing Activity in Other Experiments

The indexes (total chewing min per kilogram DMI) of this experiment's long cut,
no WFCS diet (34 min) and short cut, no WFCS diet (31 min) were higher than the index
of 26 min/kg DMI for "medium cut alfalfa silage" as reported in review by Sudweeks et
al. (58) -- even though the NDF contents of the diets on this experiment were probably
lower than the NDF content of the only silage diet.

Grant et al. (31) had alfalfa silages that were similar in composition (40% NDF)
and in proportion of diet (55% of diet DM) as this experiment. This experiment's long
cut, no WFCS diet recorded total chewing activity results similar to their "coarse" diet
' (34.3 min per kg DMI vs. 33.1 min per kg DMI and 116.2 min per kg NDF and 116.5
min per kg NDF). This experiment's short cut, no WFCS diet total chewing activities

(30.8 min per kg DMI and 100.0 per kg NDF) was closer to their "medium" cut (30.5 per

73
kg DMI min and 107.7 min kg NDF) than their "ﬁne" cut (25.4 min per kg DMI and 91.0
min kg NDF).

Clark and Armentano (15) also used an alfalfa silage. The NDF content in that
experiment was 43% -- one percent higher than this experiment. Their high alfalfa silage
control diet had 757 min of total daily chewing activity and 4.3 kg of corrected
(nongrain) NDF intake. This experiment's long cut, no WFCS diet had 785 min of total
daily chewing activity and 5.3 kg of corrected NDF intake and this experiment's short cut,
no WFCS diet had 697 min and 5.5 kg, respectively. If total minutes per kilogram
corrected NDF intake is calculated, their high alfalfa silage control diet is more effective
in stimulating chewing per kilogram than this experiment's long cut, no WFCS diet and
short cut, no WFCS diet (176 min per kg corrected NDF vs. 150 min per kg corrected
NDF and 128 min per kg corrected NDF, respectively).

eNDF Coefficients for This Experiment

Based on the calculated ﬁber effectiveness coefﬁcients, the long cut alfalfa silage
NDF stimulated chewing activity more than the short cut alfalfa silage NDF. The two
silages in turn had greater effectiveness of ﬁber than WFCS. Greater chewing activity
probably increased saliva ﬂow and probably increased buffer secretion into the rumen
which neutralized greater amount of fermentation acids.

Fiber effectiveness results in this experiment suggest three things: (1) that WFCS
is less effective than alfalfa silage as measured by total chewing activity per kilogram
corrected NDF intake, (2) results of by—product eNDF calculations are dependent on the
character of the background forage, and (3) effectiveness measurements are similar when
calculated with either total chews per kilogram corrected NDF intake or total chewing
time per kilogram corrected NDF intake as measured by an electronic collection system
(18) in a recommended experimental design (19). WFCS NDF is a fraction of the

effectiveness of alfalfa silage NDF depending on silage particle size. The two silages

74
were of particle sizes that are typical for Michigan dairy farms. While milk fat
percentage were similar across experimental diets, chewing activity differences were
detected. How critical these differences are would need to be determined by further

study.

Comparisons to Effectiveness Coefficients of Other Experiments

In the only other study to quantify whole WFCS effectiveness, Clark and
Armentano (15) measured ﬁber effectiveness as the ability to correct milk fat depression.
Using their slope coefﬁcient method, they calculated .80 for dried distiller's grains in
alfalfa silage and 1.30 for WFCS in alfalfa silage. These calculations suggest dried
distiller's grains NDF is 20% less effective than alfalfa silage NDF and that WFCS NDF
is a third more effective than alfalfa silage NDF. However, there are several concerns of
using the slope coefﬁcient approach with milk fat contents in this experiment. First, milk
fat percentages across all diets were not different (P>.05). They suggested, that while
accurate, the fat test method lacked precision and was therefore unable to separate means.
However, an alternate explanation is that, due to the nonsigniﬁcance, all milk fat
percentages are a sampling of the same mean. If is true, then, as they noted, dried
distiller's grains and WFCS are approximately the same as alfalfa silage in ﬁber
effectiveness (i.e. all = 1.0) as measured by ability to stimulate fat synthesis.

Since milk fat percentage is a result of the whole diet intake, two other concerns
are noted. First, the high alfalfa silage diets had the lowest DMI (P<.01) and lowest diet
NEL concentrations as calculated from NRC values. These two factors combined to give
the high alfalfa silage diets an intake of 3 less Mcal per day than all other diets. Second,
WFCS alters milk fat fatty acid proportions as discussed earlier. The net result of these
two factors on milk fat percentage is unknown.

If the results for total chewing time reported by Clark and Armentano are put into

the effectiveness equations developed in this experiment, the effectiveness of dried

75
distiller's grain NDF in alfalfa silage is .60 ((155-(.70)(176))/((.30)(176)) and the
effectiveness of WFCS NDF in alfalfa silage is .88 ((l67-(.56)(176))/((.44)(176)) based
on total chewing time per kilogram corrected NDF intake. In both cases, effectiveness
coefﬁcients less than those calculated with the milk fat methods and less than alfalfa
silage NDF.

Comparison of Methods of Effective Fiber Calculation

The assumptions of this experiment's method of effective ﬁber calculation is
compared graphically to the method used at University of Wisconsin in Figures 4 and 5,
respectively. In both systems, corrected (nongrain) NDF intake is used because the grain
contribution to ﬁber effectiveness is assumed to be zero. This assumption is based on a
review of the literature. This experiment's calculations are based one assumption, two
data points and one measured proportion. Starting with the assumption that there is no
chews at zero kilograms corrected NDF intake, chews increase linearly with corrected
NDF intake to the measured total chews at the measured kilogram corrected NDF intake
of the no WFCS diet. Next, the breakpoint for WFCS NDF is determined by using the
measured proportion of WFCS NDF replacing alfalfa silage NDF to partition out the
NDF intake and chews due to alfalfa silage NDF. From this breakpoint a line is drawn to
measured total chews by kilogram corrected NDF intake of added WFCS diet.
Effectiveness calculation is equivalent to the slope of breakpoint to added WFCS diet
data point divided by slope of origin to the no WFCS diet data point.

The calculations of University of Wisconsin are shown in Figure 5. The axes of
Figure 5 change from Figure 4. The X axis is corrected NDF content of diet showing this
method does not account for intake. This is problem if treatment affects DMI. The Y
axis is their measure of effectiveness, milk fat content. The milk fat content is measured
on three diets to obtain the two slepes needed for the ﬁber effectiveness calculation.

From the milk fat content on the low alfalfa silage NDF diet, two lines are drawn to the

76
milk fat contents of high ﬁber diets -- one to the high ﬁber alfalfa silage control and one
to the test by-product. A linear response of milk fat content from low ﬁber to high ﬁber
is assumed. Again, they negate grain NDF in their calculation of NDF content as they
assume grain has no effect on milk fat content. With the use of low fiber control, their
slopes are not forced through the origin meaning that, at zero NDF content of diet, milk

fat percentage does not necessarily equal to zero.

 

‘ V +——>
Proportion of Added wrcs Diet Proportion of Added WFCS Diet
NDF from Alfalfa Silage NDF from WFCS

   
  
 

Breakpoint

Total Chews per Day

 

O No WFCS Diet
0 Added WFCS Diet

 

 

 

 

Corrected NDF Intake (kg/day)

Figure 4. Graphical representation of assumptions and mathematics of this study. The
calculation of ﬁber effectiveness is the slope "B" divided by slope "A."

77

 

0 Low NDF Control
0 High NDF Control
x Test NDF

   

 

 

 

Milk Fat Percentage

 

 

l 1
LOW HIGH

Corrected NDF Content of Diet

Figure 5. Graphical representation of assumptions and mathematics of the slope
coefﬁcient method used by Armentano and coworkers (14, 15, 61, 67). The calculation
of ﬁber effectiveness is the slope "B" divided by slope "A."

Assumptions of Calculations of eNDF for This Experiment

The ﬁrst assumption was that the contribution of the grain mix NDF to total
chewing activity measured on experimental diets was negligible. Grain mix NDF was
subtracted from total diet NDF to yield corrected NDF intake. Ruminants have a very
strong drive to ruminate (69, 70). Rumination may provide the psychological benefits of
sleep as well (1). On grain only and pelleted diets, ruminating chews are low in number
and irregular in nature (6). These three points taken together suggest that ruminating

chews on these diets may be as much higher brain stimulated as ruminally stimulated. If

78

a coarse forage were added to these diets, one would expect the stimulus of forage to
ellipse, not add, to any stimulation already present because of the forage's greater ability
to stimulate rumination.

The second assumption was that the total chewing activity responds linearly to
NDF intake across the range of intakes measured in this experiment. With the signiﬁcant
effects on DMI, it was necessary to correct total chews and total time chewing for intakes
to avoid the bias of intake. A calculation to test for linearity using the results of the
effectiveness calculations supports this assumption. If the eNDF coefﬁcient of WFCS in
short cut alfalfa silage is multiplied by the short cut to long cut ratio, the product should
equal the eNDF coefﬁcient of WFCS in long cut alfalfa silage if there is linear
relationship. Using the ranges for coefﬁcients calculated with total chews per kilogram
corrected NDF intake [(.93 to .67) .1- (.86 to .81) = (.58 to .38)] and with total chewing
time per kilogram corrected NDF intake [(.84 to .64) * (.88 to .84) = (.61 to .44)] suggest
that linearity is a reasonable assumption. In the future, these assumptions will have to be

tested.

Choice of Method of eNDF Calculations for This Experiment

The presence of sufficient effective ﬁber in the diets has a number of effects. An
individual animal consumes the ﬁber of the diet and responds according to the chemical
and physical properties of the ﬁber. Higher diet proportions of ﬁber and coarser texture
of ﬁber stimulate chewing activity increasing saliva ﬂow. Increased saliva ﬂow adds
more bicarbonate to the rumen. Higher NDF proportions of the diet promote an acetate
fermentation and maintain a more neutral ruminal pH. This fermentation allows normal
milk fat percentages but, increased effective ﬁber is not linearly related to milk fat
percentage.

High acetate fermentation and high chewing stimulus are not always linked. An
example of this disassociation is soyhulls. Soyhulls are high in ferrnentable ﬁber and

their substitution for grain may increase milk fat due to increased acetate production yet

79
may not provide any buffering due to salivation because of their lower ability to stimulate
chewing.

Total chews per kilogram corrected NDF intake, chosen as the measurement of
ﬁber effectiveness for this experiment, has several advantages. First, lactation is not
required and, therefore, may be tested and applied over a range of ruminants. Second, the
measurement of chewing activity and intake is non invasive as it does not require the
rumen cannulation necessary for accurate sampling of ruminal acids and pH. And third,
the measurement of chewing activity is a more direct measure of ﬁber effectiveness,
rumen health, and rumen action than milk fat percentages or ruminal acid ratios.

The use of total chews per kilogram corrected NDF intake to calculate
effectiveness of ﬁber was an extension of Balch's index (6), total chewing time per
kilogram DMI. Total chews were chosen over total chewing time because (1) the EDC
system used (18) allowed easy and accurate measurement of individual chews and (2)
saliva output may be more related to activity than time. However, indexes as measured
with total chews and total time were determined for later comparison. NDF intake was
chosen to partition total chews rather than DMI because forages and by-products vary so
greatly in NDF.

Effectiveness coefﬁcients were calculated using total chewing activity, the sum of
eating and ruminating chewing activity. As noted by Erdman (25), saliva ﬂow rates are
greater during rumination than during eating. In the future, more accurate eNDF
coefﬁcients might be obtained by dividing total chewing activity back into its component
parts, ruminating activity and eating activity, with each adjusted by a coefﬁcient
describing saliva ﬂow.

Enough effective ﬁber in the diet allows commonly reported (mid 3's) milk fat
percentages. However, milk fat percentage may not be the best measure of ﬁber
effectiveness particularly with WFCS in the diet. Milk fat percentage, an indirect

indicator, is a function of several factors and milk fat percentage and level of dietary ﬁber

80
are not highly related. Regressions of increased ﬁber in diet, increased ruminal A:P, and
increased milk fat percentage are related by linear components but, the relationships are
still not direct (25). Milk fat percentage results may not be solely a response to ﬁber
effectiveness, especially in early lactation when milk fat is less responsive to ﬁber levels
in the diet due to the high adipose mobilization during the negative energy balance.

Fatty acids of milk fat triglycerides come from two sources. Triglyceride fatty
acid chains of 16 carbons or less are synthesized in mammary gland from acetate and beta
hydroxybutyrate. Triglyceride fatty acid chains of 18 carbons or more are transferred
from the diet and from the adipose tissue. Though often equal contributors to the total
milk fat produced, the balance of the two processes is a function of such factors as
inﬂuence of diet components, character of rumen fermentation and energy balance (65).
Measuring ﬁber effectiveness of diets containing WFCS or WFCS products against diets
containing no WFCS using milk fat percentage as an indicator may have dubious utility
as WFCS alters fatty acid composition of milk fat (34, 54, 55) as changes in milk fat
percentage may not be the result of ﬁber effectiveness.

Milk fat percentage is also becoming less important to the producer. Due to less
consumer demand, milk fat has been and will continue to become less important
economically. Producers will be increasing ruminally fermented carbohydrate in the cow
diet to increase ﬂuid milk and milk protein production. This change will shift the
emphasis of effective ﬁber from its effect on milk fat percent to its effect on buffering the

fermentation acids produced by increased ruminally fermented carbohydrate.

SUMMARY

Dry matter intake, NDF intake, and water intake were increased by short length of
cut and with WFCS addition. Milk production was increased with the addition of WFCS
to treatment diets. Milk composition was not affected by treatment. Chewing activities
as measured by count and time were reduced by the short length of cut and by WFCS
addition. Total chews per kilogram NDF intake were reduced with short length of cut
and with WFCS addition. Total chews per kilogram NDF intake were used to calculate
the effectiveness of WFCS NDF. The interaction of length of cut and WFCS treatment
caused two different coefﬁcients for effectiveness of WFCS NDF. Effectiveness of
WFCS NDF in long cut alfalfa silage was calculated as .48 i .10 and effectiveness of
WFCS NDF in short cut alfalfa silage was calculated as .80 i .13.

81

CONCLUSIONS

Measurement of by-product effectiveness is inﬂuenced by the background forage
the by-product is measured against and, therefore, characterization of background forage
is extremely important. Using total chews per kilogram corrected NDF intake to measure
effectiveness of by-product NDF has merit but, further research is needed. Effective
NDF of whole fuzzy cottonseed as measured by chews per kilogram corrected NDF

intake ranges from approximately 50% of long cut alfalfa silage NDF to approximately
80% of short cut alfalfa NDF.

82

FUTURE WORK

This experiment suggests that the use of total chewing activity per kilogram
corrected NDF intake is viable. However, the assumptions used in this experiment need
to be tested: Is the relationship between total chews and NDF intake linear?, Are the
chews from grain mix proportion of the diet negligible?, and Is chews per kilogram of
WFCS NDF constant at different diet proportions? With the total chewing activity per
kilogram corrected NDF intake system, critical, minimum, and maximum levels need to
be determined for effectiveness of ﬁber.

In the future, more accurate eNDF coefﬁcients might be obtained by dividing total
chewing activity back into its component parts, ruminating activity and eating activity,
with each adjusted by a coefﬁcient describing saliva ﬂow. As noted by Erdman (25),
saliva flow rates are greater during rumination than during eating.

When testing by-product ﬁber effectiveness, by-product is fed with a background
forage with distinct characteristics. It is possible that the forage fed with the by-product
is variable in ﬁber effectiveness from experiment to experiment. These differences may
inﬂuence the measurement of the by-product effectiveness. Therefore, the background
forage must be characterized as in the past and the characterization reported, particularly
particle size.

A cooperative effort among research groups for more standardization of reporting
of experimental parameters would be beneﬁcial. Reporting a standard set of variables is
needed for comparison of effective ﬁber systems until one approach proves superior.

When testing by-product effectiveness, reporting the characterization of background

83

84
forages remains important. Also, across all effectiveness studies, a standard of
comparison needs to be agreed upon. Whether the standard is coarse alfalfa silage, long
grass hay, 100% NDF forage extrapolated from long grass hay, or some other, remains to

be determined.

APPENDICES

Appendix A. Cow days lost for 92CSM1 and reason they were removed from ﬁnal data

set.

85

 

removed from

 

 

Cow Period Day Reason removed chewing data drinking data
863 3 5 halter failure x

2347 2 5 halter failure x

2473 l l halter failure x

2480 l 2 halter failure x

2480 3 1 halter ﬁt too tight x

2502 1 l halter failure x

2502 1 2 halter failure x

2552 l 5 uncorrectable interference x

2552 2 2 health; suspected DA x x
2552 2 3 health; suspected DA x x
2552 2 4 health; suspected DA x x
2552 2 5 health; conﬁrmed DA x x
2616 l 5 halter failure x

2616 4 l uncorrectable interference x

2616 4 2 halter failure x

2617 2 1 uncorrectable interference x

2619 2 5 halter failure x

2619 3 2 halter failure x

2619 4 1 halter failure x

2635 2 3 possible halter failure x

 

DA = displaced abomasum

86
Appendix B. Pro-experiment individual cow information for 92CSM1.

 

 

Stall cow FDATE lact. # Square 305MB DIM BW Milk #
1 2347 6-24-92 4 1 22670 77 640 97
2 2525 5-24-92 2 1 25800 108 677 103
3 2552 5- 8-92 2 1 27140 124 597 100
4 2473 6—14-92 3 1 23190 87 599 87
5 2502 5- 2-92 2 2 28340 130 590 100
6 863 4- 4-92 2 2 24120 158 571 83
7 535 4-17-92 6 2 23960 145 523 100
8 2480 4- 3-92 3 2 26980 159 710 96
9 2619 5-23-92 1 3 27660 109 593 86

10 2617 3-21-92 1 3 27420 172 556 80
11 2616 4-11-92 1 3 23170 151 556 73
12 2635 6-17-92 1 3 26270 84 501 79

 

FDATE = date of freshening

lact. # = lactation number

305MB = current projected 305 d mature equivalent milk production

DIM = days in milk on 9-8-92

BW = initial body weight in kilograms

Milk # = pre-experiment daily milk production in pounds

87

Appendix C. Summary of cow information by square on Sept. 8, 1992 for 92CSM].

 

 

Square Milk DIM BW lactation

1 43.9 i 3.2 99 i 21 628 i 38 multiparous
2 43.0 i 3.7 148 i 14 598 i 79 multiparous
3 36.1 i 2.4 129 i 40 551 i 38 primiparous
ALL 41.0 i 4.6 125 i 32 593 i 60 -

 

Milk = average milk in kg (mean i standard deviation)

DIM = average days in milk (mean i standard deviation)

BW = average body weight in kg (mean 2 standard deviation)

88
Appendix D. Formulas used in 92CSM].

% solids not fat (% SNF)
= % milk protein + % milk lactose + % milk ash
-assumed .7l % for % milk ash
Source: Tyrrell and Reid. JDS 48:1215.

solids corrected milk (SCM)(lb..)
=(12.3)(fat lb.) + (6.56)(solids not fat lb.) - (.0752)(milk lb.)
Source: Tyrrell and Reid. JDS 48:1215.

kilograms = (lb.)(.454)
liters = (gallons)(3.785)

dry matter intake (DMI)
= dry matter fed (1b.) - dry matter pounds refused (1b.)
= (as fed lb.)(TMR %DM/100) - (orts lb.)(orts %DM/IOO)

% orts
= dry matter lb. refused + dry matter pounds fed

change in body condition score or body weight
= avg. @ start of the period — avg. @ the end of the period

average % fat for test week (similar for %protein, %lactose and SCC)
= (total lb. fat produced for six sampled milkings) + (total lb. of milk
produced for six sampled milkings)
= ((lb. of milk from ﬁrst milking)(ﬁrst milking % fat/ 100) + (fat lb. from

second milking) + .................. + (fat lb. from sixth milking)) + ((lb. of milk
from ﬁrst milking) + (lb. of milk from second milking) + .................... + (lb
of milk from sixth milking))

4.0% FCM (source: NRC)
= milk kg (.4 + (.15)(% milk fat))

3.5% FCM (source: NRC)
= milk kg (.43 + (.l62)(% milk fat))

LIST OF REFERENCES

10

ll

12

LIST OF REFERENCES

Albright, J. L. 1993. Feeding behavior in dairy cattle. J. Dairy Sci. 76:485.

Allen, M. S., J. B. Robertson, and P. J. VanSoest. 1984. Acomparison of particle
size methodologies and statistical treatments. Page 39m Techniques 1n particle
size analysis of feed and digesta 1n ruminants. Proc. Workshop Can. Soc. Anim.
Sci., Banff. P. M. Kennedy,ed. Occas. Publ. 1, Can. Soc. Anim. Sci. Edmonton,
Alta., Can.

Association of Official Analytical Chemists. 1984. Ofﬁcial methods of analysis.
14th Ed. AOAC, Washington, DC.

Bae, D. H., J. G. Welch, and B. E. Gilman. 1983. Mastication and rumination in
relation to body size of cattle. J. Dairy Sci. 66:2137.

Bae, D. H., J. G. Welch, and A. M. Smith. 1981. Efﬁciency of mastication in
relation to hay intake by cattle. J. Anim. Sci. 52:1371.

Balch, C. C. 1971. Proposal to use time spent chewing as an index of the extent
to which diets for ruminants possess the physical property of ﬁbrousness
characteristic of roughages. Br. J. Nutr. 26:383.

Ballesteros, J. P. G. 1975. The effects of the chopped of the ensilaged alfalfa

and conservation system upom alimentary behaviour of sheep-[sic] Revista Nutr.
Anim. 13:7.

Beauchemin, K. A. 1991. Ingestion and mastication of feed by dairy cattle.
Veterinary Clinics of North America: Food Animal Practice. 7(2):439.

Beauchemin, K. A. 1991. Effects of dietary neutral detergent ﬁber concentration
and alfalfa hay quality on chewing, rumen function, and milk production of dairy
cows. J. Dairy Sci. 74:3140.

Beauchemin, K. A., and J. G. Buchanan-Smith. 1989. Effects of dietary neutral
detergent ﬁber concentration and supplementary long hay on chewing activities
and milk production of dairy cows. J. Dairy Sci. 72:2288.

Beauchemin, K. A., B. I. Farr, and L. M. Rode. 1991. Enhancement of the
effective ﬁber content of barley-based concentrates fed to dairy cows. J. Dairy
Sci. 74:3128.

Cassida, K. A., and M. R. Stokes. 1986. Eating and resting salivation in early
lactation dairy cows. J. Dairy Sci. 69:1282.

89

13

14

15

16

17

18

19

20

21

22

23

24

25

26

90

Church, DC. 1988. Salivary function and production. Page 117 in The ruminant
animal: digestive physiology and nutrition. D. C. Church, ed. Prentice Hall,
Englewood Cliffs, NJ.

Clark, P. W. and L. E. Armentano. 1993. Replacement of alfalfa NDF with a
combination of nonforage ﬁber sources. J. Dairy Sci. 76(1):210. (Abstr)

Clark, P. W. and L. E. Armentano. 1993. Effectiveness of neutral detergent ﬁber
in whole cottonseed and dried distillers grain compared with alfalfa haylage. J.
Dairy Sci. 76:2644.

Coppock, C. E., J. K. Lanham, and J. L. Hornet. 1987. A review of the nutritive
value and utilization of whole cottonseed, cottonseed meal and associated by-
products by dairy cattle. Anirn. Feed Sci. Tech. 18:89.

Coppock, C. E., and D. L. Wilks. 1991. Feeding whole cottonseed and cottonseed
meal to dairy and beef cattle. Page 43 In Proceedings of National Invitational
Symposium: Alternative feeds for dairy and beef cattle. St. Louis, MO.

Dado, R. G., and M. S. Allen. 1993. Continuous computer acquisition of feed
and water intakes, chewing, reticular motility, and ruminal pH of cattle. J. Dairy
Sci. 76:1589.

Dado, R. G., and M. S. Allen. 1994. Variation in and relationships among
feeding behavior variables measured continuously from lactating dairy cows. J.
Dairy Sci. 77:132.

DeBoever, J. L., J. l. Andries, D. L. DeBrabander, B. G. Cottyn, and F. X.
Buysse. 1990. Chewing activity of ruminants as a measure of physical structure
-- a review of factors affecting it. Anim. Feed Sci. Tech. 27:281.

DeBoever, J. L., D. L. DeBrabander, A. M. DeSmet, J. M. Vanacker, and C. V.

Boucque. 1993. Evaluation of physical structure. 2. maize silage. J. Dairy Sci.
76:1624.

DeBoever, J. L., A. DeSmet, D. L. DeBrabander, and C. V. Boucque. 1993.
Evaluation of physical structure. 1. grass silage. J. Dairy Sci. 76:140.

Edmonson, A. 1., I. J. Lean, L. D. Weaver, T. Farver and G. Webster. 1989. A
body condition scoring chart for Holstein dairy cows. J. Dairy Sci. 72:68.

Ehle, F. R., 1984. Measurement of mean particle size of forages by wet and dry
sieving techniques. Page 18 in Techniques in particle size analysis of feed and
digesta in ruminants. Proc. Workshop Can. Soc. Anim. Sci., Banff. P. M.
Kennedy, ed. Occas. Publ. 1, Can. Soc. Anim. Sci., Edmonton, Alta., Can.

Erdman, R. A., 1988. Dietary buffering requirements of the lactating dairy cow:
a review. J. Dairy Sci. 7123246.

Firkins, J. L. 1992. Effectiveness of ﬁber from non-forage sources. Page 7 in
Proc. of T ri-State Dairy Nutrition Conf. Fort Wayne, IN.

27

28

29

30

31

32

33

34

35

36

37

38

39

41

91

Firkins, J. L., and M. L. Eastridge. 1992. Replacement of forage or concentrate
with combinations of soyhulls, sodium bicarbonate, or fat for lactating dairy
cows. J. Dairy Sci. 75:2752.

Goering, H. K., and P. J. VanSoest. 1970. Forage ﬁber analysis (apparatus,

reagents, procedures, and some applications). Agric. Handbook No. 379. ARS-
USDA, Washington, DC.

Grant, R. J ., V. F. Colenbrander, and J. L. Albright. 1990. Effect of particle size
of forage and rumen cannulation upon chewing activity and laterality in dairy
cow. J. Dairy Sci. 73:3158.

Grant, R. J ., V. F. Colenbrander, and D. R. Mertens. 1990. Milk fat depression in
dairy cows: role of particle size of alfalfa hay. J. Dairy Sci. 73: 1823.

Grant, R. J., V. F. Colenbrander, and D. R. Mertens. 1990. Milk fat depression in
dairy cows: role of silage particle size. J. Dairy Sci. 73:1834.

Hach, C. C., B. K. Bowden, A. B. Lopelove, and S. V. Brayton. 1987. More
powerful peroxide kjeldahl digestion method. J. Assoc. Ofﬁc. Anal. Chem.
70(5):783.

Harris, B. 1991. Value of high-ﬁber alternative feedstuffs as extenders of
roughage sources. Page 138 In Proceedings of National Invitational Symposium:
Alternative feeds for dairy and beef cattle. St. Louis, MO.

Hawkins, G.E., K. A. Cummins, M. Silverio, and J. J. Jilek. 1985. Physiological
effects of whole cottonseed in the diet of lactating dairy cows. J. Dairy Sci.
68:2608.

Hsu, J. T., D. B. Faulkner, K. A. Garleb, R. A. Barclay, G. C. Fahey, Jr., and L. L.
Berger. 1987. Evaluation of corn ﬁber, cottonseed hulls, oat bulls and soybean
hulls as roughage sources for ruminants. J. Anim. Sci. 65:244.

Jaster, E. H., and M. R. Murphy. 1983. Effects of varying particle size of forage
on digestion and chewing behavior of dairy heifers. J. Dairy Sci. 66:802.

JMP® User's Guide: Software for statistical visualization on the Apple®
Macintosh®. Version 2 Edition. 1989. SAS Inst, Inc., Cary, NC.

Jorgensen, N. A., M. F. Finner, and J. P. Marquardt. 1978. Effect of forage
particle size on animal performance. Paper No. 78-1048. Am. Soc. Agric.
Engineers. St. Joseph, MI.

Kalu, B. A., and G. W. Fick. 1981. Quantifying morphological development of
alfalfa for studies of herbage quality. Crop Sci. 21:267.

Kick, C. H., P. Gerlaugh, A. F. Schalk, and E. A. Silver. 1937. The effect of
mechanical processing of feeds on the mastication and rumination of steers. J.
Agric. Res. 55(8):587.

Krishnamoorthy, U., T. V. Muscato, C. J. Sniffen, and P. J. VanSoest. 1982.
Nitrogen fractions in selected feedstuffs. J. Dairy Sci. 65:217.

42

43

45

46

47

48

49

50

51

52

53

54

55

56

57

92

Lofgren, P. A. and R. G. Warner. 1970. Inﬂuence of various ﬁber sources and
fractions on milk fat percentage. J. Dairy Sci. 53: 296.

McDonald, P. 1981. The biochemistry of silage. John Wiley & Sons, Ltd.,
Chichester, England.

Mertens, D. R. 1980. Fiber content and nutrient density in dairy rations. Proc.
Distillers Feed Conf. 35:35.

Mertens, D. R. 1991. Physical and chemical characteristics of ﬁber that determine
its roughage value. Proc. of the 17th Biennial Ruminant Health-Nutrition Conf.
of NY 8 Vet Med. Society: Feeding and managing cows in transition.

National Research Council, 1989. Nutrient requirements of dairy cattle. 6th rev.
ed. Nat. Acad. Sci., Washington, DC.

Norgaard, P. 1989. The inﬂuence of physical form of ration on chewing activity
and rumen motility in lactating cattle. Acta Agric. Scand. 39:187.

Okamoto, M., K. Deoka, and T. Bando. 1979. Effects of cut lengths of corn
silage on ruminating behavior of cows. Bull. No. 10 Shintoku Anim. Husb. Exp.
Stn.

Pitt, R. E. 1990. Silage and hay preservation (NRAES-S). Northeast Regional
Agricultural Engineering Service, Ithaca, NY.

Santini, F. J., A. R. Hardie, N. A. Jorgensen, and M. F. Finner. 1983. Proposed
use of adjusted intake based on forage particle length for calculation of roughage
indexes. J. Dairy Sci. 66:811.

Sarwar, M., J. L. Firkins, and M. L. Eastridge. 1991. Effect of replacing neutral
detergent ﬁber of forage with soyhulls and corn gluten feed for dairy heifers. J.
Dairy Sci. 74:1006.

Shaver, R. D. 1990. Forage particle length in dairy rations. Page 58 in Proc.
Dairy Feeding System Symp., Harrisburg, PA.

Shaver, R. D., A. J. Nytes, L. D. Satter, and N. A. Jorgensen. 1986. Inﬂuence of
amount of feed intake and forage physical form on digestion and passage of
prebloom alfalfa hay in dairy cows. J. Dairy Sci. 69: 1545.

Smith, N. E., L. S. Collar, D. L. Bath, W. L. Dunkley, and A. A. Franke. 1981.
Digestibility and effects of whole cottonseed fed to lactating cows. J. Dairy Sci.
64:2209.

Smith, W. A., B. Harris, Jr., H. H. Van Horn, and C. J. Wilcox. 1993. Effects of
forage type on production of dairy cows supplemented with whole cottonseed,
tallow, and yeast J. Dairy Sci. 76:205.

Stern, M. D. and C. J. Ziemer. 1993. Consider value, cost when selecting
nonforage ﬁber. Feedstuffs. 65(2): 14.

Sudweeks, E. M. 1977. Chewing time, rumen fermentation and their relationship
in steers as affected by diet composition. J. Anim. Sci. 44:694.

58

59

60

61

62

63

64

65

66

67

68

69

70

71

93

Sudweeks, E. M., L. O. Ely, D. R. Mertens, and L. R. Sisk. 1981. Assessing
minimum amounts and form of roughages in ruminant diets: roughage value
index system. J. Anim. Sci. 53:1406.

Sudweeks, E. M., L. O. Ely, and L. R. Sisk. 1980. Effect of intake on chewing
activity of steers. J. Dairy Sci. 63:152.

Sudweeks, E. M., S. E. Law, L. O. Ely, M. E. McCullough, and L. R. Sisk. 1979.
Development and application of a roughage value index for formulating dairy
rations. Bulletin No. 238 Georgia Agric. Exp. Stn.

Swain, S. M., K. K. Vaughan, and L. E. Armentano. 1991. Determining

effectiveness of ﬁber in by-product feeds compared to alfalfa silage. J. Dairy Sci.
74( 1):220.(Abstr.)

Tyrrell, H. F. and J. T. Reid. 1965. Prediction of the energy value of cow's milk.
J. Dairy Sci. 48:1215.

VandeHaar, M., H. Bucholtz, R. Beverly, R. Emery, M. Allen, C. Sniffen, and R.
Black. 1992. Spartan dairy ration evaluator/balancer: an agricultural integrated
management software microcomputer program. Michigan State Univ.
Cooperative Ext, East Lansing, MI.

VanHom, H. H., B. Harris, Jr., M. J. Taylor, K. C. Bachman, and C. J. Wilcox.
1984. By—product feeds for lactating dairy cows: effects of cottonseed hulls,
sunﬂower hulls, corrugated paper, peanut hulls, sugarcane bagasse, and whole
cottonseed with additives of fat, sodium bicarbonate, and Aspergillus oryzae
product on milk production. J. Dairy Sci. 67:2922.

VanSoest, P. J. 1982. Nutritional Ecology of the Ruminant. Comstock Publishing
Associates, Ithaca, NY.

VanSoest, P. J ., J. B. Robertson, and B. A. Lewis. 1991. Methods for dietary
ﬁber, neutral detergent ﬁber and nonstarch polysaccharides in relation to animal
nutrition. J. Dairy Sci. 74:3583.

Vaughan, K. K., S. M. Swain, and L. E. Armentano. 1991. Effectiveness of
NDF from ground corn cobs and wheat middlings compared to alfalfa silage. J.
Dairy Sci. 74(1):220.(Abstr.)

Wagner, K. M., J. L. Firkins, M. L. Eastridge, and B. L. Hull. 1993.
Replacement of corn silage with wheat middlings and calcium chloride or sodium
bicarbonate for lactating dairy cows. J. Dairy Sci. 76:564.

Welch, J. G. 1982. Rumination, particle size and passage from the rumen. J.
Anim. Sci. 54:885.

Welch, J. G., and A. P. Hooper. 1988. Ingestion of feed and water. Page 108 in
The ruminant animal: digestive physiology and nutrition. D. C. Church, ed.
Prentice Hall, Englewood Cliffs, NJ.

Welch, J. G., and A. M. Smith. 1969. Inﬂuence of forage quality on rumination
time in sheep. J Anim. Sci. 28:813.

72

73

74

75

76

77

94

Welch, J. G., and A. M. Smith. 1969. Effect of varying amounts of forage intake
on rumination. J. Anim. Sci. 28:827.

Welch, J. G., and A. M. Smith. 1970. Forage quality and rumination time in
cattle. J. Dairy Sci. 53:797.

Welch, J. G., and A. M. Smith. 1971. Physical stimulation of rumination activity.
J. Anim. Sci. 33:1118.

Welch, J. G., A. M. Smith, and K. S. Gibson. 1970. Rumination time in four
breeds of dairy cattle. J. Dairy Sci. 53:89.

Woodford, J. A., N. A. Jorgensen, and G. P. Barrington. 1986. Impact of dietary
ﬁber and physical form on performance of lactating dairy cows. J. Dairy Sci.
69: 1035.

Woodford, S. T., and M. R. Murphy. 1988. Effect of forage physical form on
chewing activity, dry matter intake, and rumen function of dairy cows in early
lactation. J. Dairy Sci. 71:674.