Citation
Effects of Dietary Physical Form and Carbohydrate Profile on Captive Giraffe

Material Information

Title:
Effects of Dietary Physical Form and Carbohydrate Profile on Captive Giraffe
Creator:
KEARNEY, CELESTE C. ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Alfalfa ( jstor )
Fermentation ( jstor )
Forage ( jstor )
Mastication ( jstor )
Nutrients ( jstor )
Particle size classes ( jstor )
Ruminants ( jstor )
Starches ( jstor )
Sugars ( jstor )
Zoos ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Celeste C. Kearney. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
5/1/2005
Resource Identifier:
71279536 ( OCLC )

Downloads

This item has the following downloads:

kearney_c ( .pdf )

kearney_c_Page_094.txt

kearney_c_Page_014.txt

kearney_c_Page_063.txt

kearney_c_Page_098.txt

kearney_c_Page_007.txt

kearney_c_Page_059.txt

kearney_c_Page_032.txt

kearney_c_Page_054.txt

kearney_c_Page_042.txt

kearney_c_Page_075.txt

kearney_c_Page_008.txt

kearney_c_Page_102.txt

kearney_c_Page_037.txt

kearney_c_Page_078.txt

kearney_c_Page_076.txt

kearney_c_Page_106.txt

kearney_c_Page_084.txt

kearney_c_Page_024.txt

kearney_c_Page_083.txt

kearney_c_Page_065.txt

kearney_c_Page_022.txt

kearney_c_Page_013.txt

kearney_c_Page_029.txt

kearney_c_Page_104.txt

kearney_c_Page_001.txt

kearney_c_Page_047.txt

kearney_c_Page_025.txt

kearney_c_Page_015.txt

kearney_c_Page_009.txt

kearney_c_Page_050.txt

kearney_c_Page_026.txt

kearney_c_Page_092.txt

kearney_c_Page_021.txt

kearney_c_Page_018.txt

kearney_c_Page_072.txt

kearney_c_Page_062.txt

kearney_c_Page_033.txt

kearney_c_Page_060.txt

kearney_c_Page_100.txt

kearney_c_Page_101.txt

kearney_c_Page_088.txt

kearney_c_Page_010.txt

kearney_c_Page_044.txt

kearney_c_Page_095.txt

kearney_c_Page_067.txt

kearney_c_Page_093.txt

kearney_c_Page_038.txt

kearney_c_Page_099.txt

kearney_c_Page_045.txt

kearney_c_Page_103.txt

kearney_c_Page_073.txt

kearney_c_Page_039.txt

kearney_c_Page_020.txt

kearney_c_Page_061.txt

kearney_c_Page_043.txt

kearney_c_Page_056.txt

kearney_c_Page_091.txt

kearney_c_Page_082.txt

kearney_c_Page_105.txt

kearney_c_Page_053.txt

kearney_c_Page_086.txt

kearney_c_Page_087.txt

kearney_c_Page_046.txt

kearney_c_Page_005.txt

kearney_c_Page_048.txt

kearney_c_Page_004.txt

kearney_c_Page_023.txt

kearney_c_Page_058.txt

kearney_c_Page_036.txt

kearney_c_Page_019.txt

kearney_c_Page_027.txt

kearney_c_Page_012.txt

kearney_c_Page_055.txt

kearney_c_Page_017.txt

kearney_c_Page_074.txt

kearney_c_Page_080.txt

kearney_c_Page_089.txt

kearney_c_Page_052.txt

kearney_c_Page_049.txt

kearney_c_Page_066.txt

kearney_c_Page_030.txt

kearney_c_Page_003.txt

kearney_c_Page_002.txt

kearney_c_Page_064.txt

kearney_c_Page_051.txt

kearney_c_Page_071.txt

kearney_c_Page_096.txt

kearney_c_Page_070.txt

kearney_c_Page_028.txt

kearney_c_Page_069.txt

kearney_c_Page_035.txt

kearney_c_Page_041.txt

kearney_c_pdf.txt

kearney_c_Page_016.txt

kearney_c_Page_034.txt

kearney_c_Page_011.txt

kearney_c_Page_077.txt

kearney_c_Page_081.txt

kearney_c_Page_068.txt

kearney_c_Page_031.txt

kearney_c_Page_097.txt

kearney_c_Page_079.txt

kearney_c_Page_085.txt

kearney_c_Page_040.txt

kearney_c_Page_057.txt

kearney_c_Page_090.txt

kearney_c_Page_006.txt


Full Text












EFFECTS OF DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE ON
CAPTIVE GIRAFFE
















By

CELESTE C. KEARNEY


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Celeste C. Keamey




























This work is dedicated, with gratitude, to the Maker of all living things.

This thesis is dedicated to my family, near and far.
It is especially dedicated to
My father, who shared his deep, abiding sense of respect and wonder for all Creation
My mother, who encouraged me to pursue my goals and fight for what I believe
My siblings, natural and "adopted," who offered love and support
My nieces and nephews, a constant source of joy and pride
And most of all my loving husband,
Who made me complete.















ACKNOWLEDGMENTS

This research was a team effort. The development and execution of this study

would not have been possible without the generous contributions of numerous

individuals. My sincerest thanks go out to my committee chair, Dr. Mary Beth Hall, and

committee members, Drs. Ellen Dierenfeld, Lee McDowell, and Charles Staples, for

assistance and instruction throughout my graduate program, and for taking a chance on an

unusual project; Busch Entertainment Corporation, for permitting and sponsoring the

study; Dr. Ray Ball, Giraffe SSP Veterinary Advisor, for initial instruction in giraffe

health and nutrition, and continued selfless assistance through all phases of this research;

Dr. Ramon Littell, University of Florida Statistics Department, for assistance with

statistical design of the study; Dr. Judy St. Ledger, BEC Director of Veterinary

Pathology, for encouragement and considerable intellectual contributions; the preceding

researchers who have taken an interest in the nutrition of captive concentrate selectors,

particularly Dr. Marcus Clauss, for his literature donations and thoughtful e-mail

discussions; Heidi Bissell, Amanda Dinges and Marti Roberson, for volunteer giraffe

observations; Kellie and the Robersons, for food, housing and helpful distractions during

the study; Alexandra Amorocho, Heidi Bissell, Faith Cullens, Colleen Larson, and

especially Lucia Holtshausen and Ashley Hughes, students of the UF Dairy Nutrition

Lab, for assistance with feed mixing and laboratory analyses; John Funk, Jocelyn

Jennings, Jan Kivipelto, Sergei Sennikov and Nancy Wilkinson, UF Animal Sciences

technicians, for training and assistance with sample preparation and laboratory analyses;









and the staff of the UF Dairy Research Unit (Hague, FL), for assistance in mixing the

experimental ration. Last but certainly far from least, I would like to acknowledge and

thank the following staff at Busch Gardens, Tampa: Chris Bliss, for passing along the

knowledge and experience of generations; Chris Allen, Joaquin Alonso, Kellie Anderson,

Kathy Driggers, James Hammerton, Brian Hart, Waylon Kerr, Chris Merrifield, Charles

Moss, Jennifer Phelps, April Richardson, Pandora Sokol, Bobby Toomy and Jerry

Washburn, hoofstock keepers, for long days and late nights of hard work during the

collection phase of this labor-intensive study; Richard Baker, Alan Cross, Cindy Davis,

Andrea Demuth, Joe Devlin, Dr. John Olsen and Glenn Young, zoo and hoofstock

management, for logistical coordination and providing the essential animal, facility, and

labor resources; Dr. Mike Burton, Dr. Genevieve Demonceaux, Heather Henry, Ian

Hutchinson, Cliff Martel and Mary Port, veterinary and hospital staff, for assistance with

sample collection and analysis; and the entire Busch Gardens Zoo staff, for their ongoing

commitment to animal care, and for their thoughts, assistance and encouragement during

this project.
















TABLE OF CONTENTS

page


A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ........ .. .............. viii

LIST OF FIGURES ............................... ... ...... ... ................. .x

ABSTRACT ........ .............. ............. ...... ...................... xi

CHAPTER


1 INTRODUCTION ............... ................. ........... ................. ... .... 1

2 REVIEW OF THE LITERATURE ON DIETARY PHYSICAL FORM AND
CARBOHYDRATE PROFILE IN RUMINANT DIETS ...........................................8

P hy sical F orm .................................... ..... ............. ... .. ... ... ...............8
Effects of Particle Size on Mastication, Saliva Flow, and Ruminal pH.............9
Effects of Particle Size on Digesta Passage, Intake, and Fermentation .............13
Potential Implications of peNDF for Captive Giraffe .............. ...................18
D ietary C arbohydrate Profile........................................................... ............... 20
C arbohydrate Fractions ............................................... ............................ 20
Proxim ate analysis system ........................................ ........ ............... 21
D etergent system .... ............................................... .... ........ .......... ......21
Carbohydrates in N natural C S D iets ......................................... ............... .... 21
Effects of NFC Source on Fermentation Characteristics and Animal
Perform ance ................ ................... .... ......... .. ... ................... 24
Interaction Between Dietary Components........................................................28
Potential Implications of Dietary NFC Profile for Captive Giraffe ..................29

3 EFFECTS OF ALTERING THE PHYSICAL FORM AND CARBOHYDRATE
PROFILE OF THE DIET ON CAPTIVE GIRAFFE .............................................36

Introduction .............. ....... ....... ................... ........... ............ 36
M materials and M methods ....................................................................... ..................38
D e sig n ............. ................. .................................................................... 3 8
G iraffe ...................................................................................................3 9









F facilities .................................................................. 3 9
D ie ts .............................. .................................................................................. 4 0
Sam ple Collection and A nalyses...................................... ........ ..............41
F eedstuffs and intake........................................................... ............... 4 1
Fecal collection and analysis..................................... ......................... 43
Behavior ........................................... ..................... ..... ........ 45
Body w eight and blood sam ples.......................................... ............... 46
Statistical analysis .............................. ..... .. .. .......... .............. 47
R results and D discussion ............................. .................... .. ........ .. .............47
In ta k e .............................................................................................................. 4 8
D ig e stib ility .................................................................................................... 4 9
B eh av ior .................................................................. 5 1
B lood M measures .................................... ............. ............... .. 53
Ancillary Study Observations / Individual Animal Effects ............... ..............54
D iet S election n ...............................................................5 6
C o n c lu sio n s .................................................................................................... 5 8

APPENDIX

A INDIVIDUAL ANIMAL MEASURES .................................. 68

B CARBOHYDRATE FRACTIONING IN FEEDSTUFFS .........................................80

C INFORMATION ON CONTROL DIET COMPOSITION AND BEHAVIOR
R E C O R D IN G ........................................................................................................ 8 1

L IST O F R E FE R E N C E S ............................................................................... 84

B IO G R A PH IC A L SK E T C H ........................................................................................ 94















LIST OF TABLES


Table p

2-1 Effects of particle size of alfalfa-based dairy cow diets on chewing activity and
ru m in a l p H ...................................... ................................ ................ 3 2

2-2 Effects of concentrate level and feeding management on ruminal pH of lactating
d airy cow s. ........................................................ ................ 32

2-3 Effects of forage particle size and grain fermentability on chewing activity,
ruminal pH, and ruminal VFA profile in midlactation cows. ................................33

2-4 Nutrient intake and digestion coefficients from giraffe fed all-hay diets.................33

2-5 Chemical composition (DM basis) of five browse plants grown at Busch
Gardens in Tampa, Florida ...................................... .............................. 34

2-6 Influence of supplemental carbohydrate source fed in combination with
0.122% BW/ day of degradable intake protein on ruminal fermentation
characteristics. .........................................................................35

3-1 Design of study.................................... ................................ ........59

3-2 Chemical composition of alfalfa hay and supplements (dry matter basis) fed
to captive giraffe, and difference between supplements. .......................................60

3-3 Effects of dietary treatment on mean daily dry matter and nutrient intake,
digestion of organic matter and crude protein (apparent) and NDFOM (true),
body weight gain and body condition score ...........................................................60

3-4 Effect of diet on fecal nutrient composition (dry matter basis).............................61

3-5 Recorded behavior of captive giraffe consuming different supplements ...............62

3-6 Percentage of time female giraffe spent engaged in oral behaviors.......................62

3-7 Effects of type of dietary supplement on giraffe blood parameters. ........................62

3-8 Pearson Correlation Coefficients for correlations of blood glucose and BUN
with kilograms of nutrients consumed and digested. ..............................................64









3-9 Pearson Correlation Coefficients for correlations of blood glucose and BUN
with various blood proteins. ...... ........................... ......................................64

A-i Individual giraffe voluntary intake (DM basis)............................................ 68

A-2 Individual giraffe nutrient intake (kg) (DM basis)............... ..... ............... 69

A-3 Individual giraffe fecal output, fecal nutrient concentration (% of DM), and
apparent nutrient digestibility (% ) ..................................................... ........ ....... 70

A-4 Differences in crude protein concentration of individual fecal samples analyzed
in w et and dried form s .................. ............................ .... .... .. ........ .... 71

A-5 Minutes over 48 hours individual giraffe spent engaged in measured behaviors. ...72

A-6 Individual giraffe blood values on day 21.................................... .................74

A-7 Body weight and body condition score of individual giraffe on days 1 and 21 of
each period. .......................................... ............................ 77

B-1 NDF and NFC fractions (percent of sample DM) in feedstuffs analyzed at the
U university of Florida. ................. .................. ............... ................80

C-1 Mean analyzed chemical composition of the batches of Purina Omelene 200
(n=5) and Mazuri Browser Breeder (n=5) fed to captive giraffe during a giraffe
feeding stu dy ...................................................... .................. 8 1

C-2 Behavioral category definitions used by observers when recording giraffe
b eh av iors. ......................................................... ................ 82
















LIST OF FIGURES


Figure page

3-1 Number of minutes over 48 hours individual giraffe spent engaged in specific
and total oral stereotype behavior. ........................................ ....................... 65

3-2 Blood concentrations of non-esterified fatty acids in individual giraffe ................66

A-i Individual animal consumption of alfalfa hay and supplement as a percentage of
body weight .............. .... .......... ............... ..... ............. ........... 78

A-2 Individual animal digestion of neutral detergent fiber organic matter (NDFOM),
and apparent digestion of OM and CP. ....................................... ............... 79

B-1 Carbohydrate partitioning. (Hall, 2001) ....... ...... ................. ................. 80

C-1 Example of data sheet used to record giraffe behavior. .........................................83















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Master of Science

EFFECTS OF DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE ON
CAPTIVE GIRAFFE

By

Celeste C. Keamey

May, 2005

Chair: Mary Beth Hall
Major Department: Animal Sciences

The effects of altering physical form and carbohydrate profile of giraffe diets were

evaluated using six non-lactating adult female giraffe in a modified reversal study.

Dietary treatments consisted of a supplement ration composed of commonly fed

commercial concentrates (GF) and an experimental supplement (EF) containing greater

concentrations of sugars and soluble fiber and lesser concentrations of starch than GF, as

well as small, heavily lignified particles used to modify dietary fiber size and texture.

Each study animal was housed individually and fed EF or GF ad libitum for 21 days, and

then received the other feed supplement in the subsequent 21 day period. Alfalfa hay,

salt and water were offered ad libitum in all periods. In each period, blood samples were

collected before feeding on day 21, feed refusals and fecal samples were collected on

days 15 through 21, and behavior was recorded for 48 hr via observation and

instantaneous sampling on days 13 through 15. Feed intake, blood measures, and

minutes spent exhibiting various behaviors were evaluated.









Data were analyzed with a statistical model that included animal, period, and diet.

Data presented are least squares means. Significance was declared at P<0.10 and

tendency at 0.10
than GF. Average daily DM intake varied greatly among animals for both alfalfa hay

(0.12 to 3.94 kg/day) and supplement (2.87 to 9.26 kg/day), but did not differ between

diets. Starch intake by giraffe decreased from 0.92 kg/day on GF to 0.12 kg/ day EF,

sugar intake tended (P=0.115) to increase from 1.12 kg/day on GF to 1.53 kg/day on EF,

and neutral detergent-soluble fiber (NDSF) intake increased from 0.85 kg/day on GF to

1.19 kg/day on EF. Time engaged in supplement consumption was greater on EF than

GF and total feed consumption + rumination time tended to be greater on EF than GF,

which may have increased saliva flow and buffering of the rumen. Despite few animals

and high variability in their feed selection and intake, the data suggest that EF facilitated

small but measurable changes in animal response. Further investigation with a larger

population of animals is needed.














CHAPTER 1
INTRODUCTION

Numerous health problems that are suspected to be of nutritional origin have been

documented in captive giraffe. Pathologies that may relate to vitamin and mineral intake

or metabolism include white muscle disease (Strafuss, 1973; Burton, 1990), urolithiosis

(Wolfe et al., 2000), and dental disease (Enqvist, 2003). Pancreatic pathologies ( Fox,

1938; Fowler, 1978; Lechowski et al., 1991; Ball et al., 2002), decreased ruminal

absorptive surface area (Hofmann and Matern, 1988), ruminal acidosis (Clauss, 1998;

Clauss et al., 2002b), fermentative gastritis or rumenitis (Fox, 1938; Ball et al., 2002) and

gastrointestinal ulceration (Fox, 1938; Fowler, 1978) also have been documented. First-

year calf mortality may be as high as 45% (Lackey and LaRue, 1997). Failure of passive

transfer (Miller et al., 1996), calf mortality due to poor milk intake (Flach et al., 1997),

and anecdotal reports of calf mortality or hand-rearing due to "maternal failure" may

relate to low colostrum and milk production due to poor nutritional status of giraffe dams.

Wasting and sudden death (Fox, 1938; Chaffe, 1968; Fowler, 1978; Strandberg et al.,

1984; Junge and Bradley, 1993; Flach et al., 1997; Ball et al., 2002) are frequently

reported in the literature and anecdotally. At this time, the true proportion of captive

giraffe mortality caused by nutritional pathologies is unknown.

The term "Peracute Mortality Syndrome" (PMS) was used to describe giraffe

wasting/ sudden death by Dr. Murray Fowler in 1978, following four cases of sudden

death at the Sacramento Zoo. Concurrent disease (tuberculosis, treated with isoniazid

powder) occurred in all four giraffe; weight loss despite reportedly adequate intake and









recent (less than one month prior to death) parturition occurred in two animals.

Necropsies were performed on three of the four animals. Absence of perirenal fat and

generalized serious atrophy of adipose tissue (3 of 3 giraffe) and marked pancreatic

atrophy (2 of 3) were notable findings. In a subsequent survey, 14 of 42 responding

institutions reported unexplained deaths of giraffe or submitted necropsy reports listing

findings consistent with the Sacramento Zoo cases. Peracute death, emaciation,

concurrent disease or stress episode, serious atrophy of adipose tissue, pulmonary edema

and trachial froth, petechial hemorrhage of serosal surfaces, and gastrointestinal

ulceration were common findings.

Fowler hypothesized that both chronic predisposing factors and acute trigger

episodes contributed to the occurrence of PMS. Based on the information available at the

time, chronic protein or energy deficiencies were listed as the most likely predisposing

factors. It was recommended that giraffe be offered low fiber diets containing 15 to 18%

CP for adult non-lactating animals, and 18 to 20% CP for calves and lactating cows.

However, a 1993 follow-up study by Junge and Bradley reported nine additional cases of

PMS in giraffe offered diets meeting these protein recommendations.

Chronic energy deficiency has again come under scrutiny as a possible

predisposing factor for PMS (Ball et al., 2002). Consistent findings of depletion and

serious atrophy of adipose tissue stores (Fox, 1938; Chaffe, 1968; Strafuss and Kennedy,

1973; Fowler, 1978; Strandberg et al., 1984; Junge and Bradley, 1993; Ball et al., 2002)

are indicative of negative energy balance. When energy expenditure exceeds available

dietary energy, a catabolic state occurs, and body reserves are mobilized. Once body fat

stores are excessively depleted, rapid catabolism of even the smallest amounts of adipose









tissue takes place in an attempt to meet energy demands. The result is serious atrophy of

adipose tissue (Smith et al., 1972).

Acute energy deficiency has been proposed as an immediate cause of death in

giraffe PMS. Ball et al. (2002) reported on the rapid wasting and mortality of two female

giraffe (A and B) during the third trimester of first pregnancy. Ante- and post- mortem

findings were consistent with PMS. Histopathologic findings included serious atrophy of

mesenteric and epicardial fat, lymphohistiocytic rumenitis and pulmonary congestion in

both cases. Pancreatic atrophy and generalized muscle atrophy were also noted in giraffe

B. Blood glucose concentrations were 20 mg/dl at the time of death in giraffe A, and 3

mg/dl at 5 hours post-mortem in giraffe B. Hypoglycemia, caused by depletion of body

reserves followed by an acute stressor parturitionn), was proposed as the immediate cause

of death. Since this publication, blood samples collected within 20 minutes of death have

revealed glucose levels of 6 and 12 mg/dl in two giraffe succumbing to PMS, and 297

mg/dl at less than 1 hour post-mortem in a giraffe expiring from an observed cervical

injury (R. Ball, personal communications).

It should be noted that endemic pathologies of unknown or suspected nutritional

origin are not isolated to captive giraffe. Wasting and mortality from unknown causes,

and from digestive pathologies, have been reported in other captive concentrate selectors

(CS) (Paglia and Miller, 1992; Shochat et al., 1997; Dierenfeld et al., 2000; Clauss et al.,

2002a; Willette et al., 2002). High (30 to 40% in some collections) neonatal mortality,

wasting syndrome, ruminal hypomotility syndrome, bloat, and rumenitis have been

described as prevalent but underreported in captive duikers (Willette et al., 2002), and

hand-rearing, diet modification, and browse supplementation were among factors









associated with improved health and increased (up to 2x) lifespan in one institution

(Barnes et al., 2002). Moose, the second largest ruminant concentrate selector (CS), are

rarely exhibited in zoos because of premature mortality (Shochat et al., 1997). Moose

"wasting syndrome complex," a syndrome of suspected nutritional origin, was the

diagnosed cause of death in 47% of 131 adult mortalities (Clauss et al., 2002a). Sudden

death, frequently attributed to digestive disorders such as ruminal acidosis and bloat, is a

common occurrence in feedlot cattle (Glock and DeGroot, 1998).

While the importance of nutrition in maintaining health, welfare, and reproductive

status of captive wildlife is receiving increasing recognition, the research that can be

performed using captive exotic animals is limited by a number of factors. The number of

animals of a given species housed in a single institution is often small, and collections

typically consist of animals in different physiological states (growth, pregnancy,

lactation), making it difficult to obtain a sufficient number of similar research animals.

Since many of the species housed in zoological institutions are rare or endangered, their

conservation value prohibits placing them in potentially harmful situations. Thus, the

herd instincts and fearful or aggressive temperament of many captive ungulates limits

collection of data that requires individual housing or animal-human contact. As a result,

statistically viable data on the effects of diet modification of individually or group-fed

animals is difficult to obtain.

At the present time, in-depth nutritional knowledge of CS ruminants remains

scarce. The true nutritional requirements of giraffe and other captive CS and dietary

factors contributing to suspected nutrition-related pathologies remain unquantified.

Because of the dearth of data on the ingredient and chemical composition of diets









consumed in the wild and on nutrient requirements of exotic ruminants, domestic

ruminants have been used as models for ration formulation. However, numerous

differences between domestic and wild ruminants must be considered. While the

objective of ration formulation for most domestic ruminants is to optimize relatively

short-term production, the objective of ration formulation for captive wildlife is to

maximize longevity and long-term health and reproduction, which may extend into

decades of life. Furthermore, differences in digestive morphology and physiology may

create discrepancies in how dietary items are utilized. When ruminants are classified

according to natural diet and digestive anatomy, domestic cattle and sheep are grazers

(GR), adapted to consumption of a predominantly grass diet. The largest living ruminant,

the giraffe, is a CS, consuming primarily foliage in its natural environment. Over 40% of

the approximately 150 known extant ruminant species are classified as CS (Wood et al.,

2000) and consume little or no grass, subsisting instead on fruit and foliage from trees,

shrubs, and herbs. Differences between wild CS and domestic GR include not only

dietary constituents, but also rumen microbial population (Dehority and Odenyo, 2003)

and gastrointestinal anatomy (Hofmann, 1973, 1984).

Digestive morphophysiological differences between ruminant CS and GR have

been widely documented and discussed (Hofmann, 1973; Kay et al., 1980; Gordon and

Illius, 1996; Shipley, 1999; Ditchkoff, 2000), and a review is presented in the

proceedings of the 30th International Congress of the International Union of Physiological

Sciences (Hofmann, 1988). A brief summary of differences pertinent to the research

presented in this thesis follows. Generally, CS have proportionally larger salivary glands

than GR. Parotid glands, for example, range from 0.18 to 0.25% of body weight (BW) in









CS, but only 0.05 to 0.07% of BW in GR. The suspected increase in saliva flow in CS

may or may not facilitate increased ruminal buffering, since the apparently well-

developed ventricular groove may allow a proportion of swallowed saliva to bypass the

rumen. Such a bypass mechanism would also allow increased amounts of dietary cell

solubles to escape ruminal fermentation. The abundance of intestinal Na+/glucose co-

transporter in the brush border membrane of wild moose and roe deer is suggestive of

some mechanism of ruminal escape for dietary sugars (Rowell-Schafer et al., 1999;

Wood et al., 2000). Decreased ruminal compartmentalization and increased diameter of

the reticulo-omasal orifice in CS (Hofmann, 1973) may contribute to the observed

increased rate of digesta passage (Clauss, 1998), while the larger capacity of the lower GI

tract suggests greater reliance on hind-gut digestion. Decreased rumen capacity, greatly

increased ruminal surface area due to dense, even papillation, and increased liver size

imply a rapid rate of fermentation and nutrient absorption.

Put succinctly, these data illustrate a single point: giraffe are not cattle. Given the

known differences between GR and CS, the likelihood of unmodified domestic ruminant

feeding practices to maintain optimal health and nutritional status of captive CS appears

questionable. However, the basic biological principles of ruminant digestion and

metabolism discovered via domestic ruminant research may be able to facilitate improved

nutrition for captive CS if viewed in light of CS ruminants' unique anatomy and

physiology.

In discussing the unique anatomical arrangements of CS, Hofmann touches upon

the need for further analysis of natural foods in order to maintain captive CS under

optimal conditions. He concludes by stating: "Neglect of original conditions finally leads






7


to a failing of the delicate anatomical, physiological, biochemical-microbial balance of

the concentrate selector's ruminant stomach" (Hofmann, 1973). The research presented

in this thesis is an attempt to further the understanding of captive CS nutritional status

and requirements, and to examine possible links between dietary factors and suspected

nutritional pathologies, including PMS in captive giraffe.














CHAPTER 2
REVIEW OF THE LITERATURE ON DIETARY PHYSICAL FORM AND
CARBOHYDRATE PROFILE IN RUMINANT DIETS

Physical Form

Current feeding recommendations for zoo ungulates (Lintzenich and Ward, 1997)

do not address two of the more recent areas of focus in ruminant nutrition research:

dietary physical form and non-fiber carbohydrate (NFC) profile. The physical form of

dietary components impacts the manner in which feedstuffs are processed in the digestive

tract. Dietary physical form affects mastication (Mertens, 1997), saliva production

(Allen, 1997), ruminal development (Beharka et al., 1998), ruminal pH (Allen, 1997),

rate and extent of ruminal fermentation (Mertens, 1997), rate of digesta passage (Allen,

1996), and the proportion of unfermented nutrients passing into the lower GI tract

(Firkins, 1997; Callison et al., 2001; Yang et al., 2001). Many methods of quantifying

the physical effectiveness of feeds by evaluating dietary particle size, chewing behavior,

or milk fat production have been proposed and used in domestic ruminant research.

Physically effective fiber (peNDF) is one approach used to define the effectiveness of

dietary particle size in maintaining ruminal (and animal) health and function (Mertens,

1997). The peNDF of a feed is defined as "the product of its neutral detergent fiber

(NDF) concentration and its physical effectiveness factor," with physical effectiveness

factor determined by the ability of the feed to promote a chewing response in the animal,

as judged on a scale of 0 (not effective) to 10 (fully effective in promoting chewing). The









physical effectiveness of a feed is, in essence, a function of particle size and rate of

particle size reduction.

Effects of Particle Size on Mastication, Saliva Flow, and Ruminal pH

Little reduction in feed particle size occurs once ingesta has passed from the rumen

(Poppi et al., 1980). Chai et al. (1984) demonstrated that initial mastication and

rumination serve to reduce feed particle size. In steers fed long-stem alfalfa or brome

hays, bolus content of particles > 3.35 mm was reduced 58 to 75% by initial mastication

and 23 to 27% by rumination (Chai et al., 1984).

The ability to stimulate chewing behavior appears to vary among forage types.

Steers on high-concentrate diets spent more time (P<0.10) chewing when fed wheat straw

rather than alfalfa hay (Shain et al., 1999). The number of eating and ruminating chews

per g of DM consumed were 2.04 and 3.41 by steers fed long-stem brome hay, but only

1.26 and 1.80 by steers fed long-stem alfalfa hay, which may have been attributable to

differences in forage fragility, or rate of particle breakdown during chewing (Chai et al.,

1984).

Dietary items with a larger physical size require more time to be consumed and

generally have a greater ability to stimulate rumination. Long-stem or minimally

chopped forages, with high NDF content and long particle length, have a greater

stimulatory effect on mastication than do finely chopped forages, and a higher peNDF

value than grains or pellets (Mertens, 1997). Eating and total chewing time in lactating

Holstein cows offered a total mixed ration (TMR) increased (P<0.05) with inclusion of

additional alfalfa in the ration, and rumination and total chewing increased linearly with

increasing particle length of alfalfa silage (P<0.05) (Clark and Armentano, 2002). Time

engaged in rumination decreased (P<0.001) linearly as wheaten hay fed ad libitum to









sheep and goats was progressively switched from chopped (1 cm) to ground (3.2 mm)

and pelleted forms (McSweeney and Kennedy, 1992). Campbell et al. (1992) used ten

Hereford steers to compare the ability of five diets to stimulate chewing: A 80%

pelleted concentrate, 20% long timothy hay (control); B 80% pelleted concentrate, 20%

alfalfa cubes; C 90% pelleted concentrate, 10% alfalfa cubes; D completely pelleted

ration using corn cobs as the primary NDF source; E 80% coarse (unground grains)

concentrate, 20% long timothy hay. Time engaged in rumination and eating behaviors

are reported as minutes per gram of dry matter intake (DMI)/ BW0.75. Modifying the

physical form of concentrates had no effect on eating (P=0.702) or rumination (P=0.954)

times. Replacing timothy hay with alfalfa cubes decreased rumination from 2.58 to 1.38

to 1.47 (P=0.001) but did not affect eating behavior (P=0.897). Replacing the control

ration with the all-pelleted diet decreased eating time from 3.55 to 2.98 (P=0.063) and

rumination from 2.58 to 1.29 (P=0.001). Number of chews per gram of DMI/ BW075

also decreased on the completely pelleted ration, from 234 to 173 (P=0.005) during

eating, and from 162 to 76 (P=0.001) during rumination.

Saliva, with a pH of approximately 8.5 (Cassida and Stokes, 1986), is a primary

ruminal buffering agent, supplying approximately half of the bicarbonate entering the

rumen of domestic cattle (Owens et al., 1998). The rate of saliva flow increases during

periods of eating and rumination (Bailey, 1961). Therefore, decreasing time engaged in

chewing behavior may decrease total daily saliva production per unit of feed consumed

(Bailey, 1961; Maekawa et al., 2002).

A study by Beauchemin et al. (2003) illustrates the link between forage particle

size, chewing activity, and ruminal pH. The effects of dietary particle size on lactating









dairy cows were examined using a total mixed ration (TMR) consisting of 60% barley-

based concentrate and 40% forage (DM basis). Forage consisted of alfalfa silage (AS)

and alfalfa hay (AH) in a 50:50 or 25:75 ratio. Alfalfa hay was coarsely chopped (CH) or

ground (GH) through a 4mm screen. Mean dietary particle length (MPL) was highest for

the TMR containing 50:50 AS:CH, followed by 25:75 AS:CH, 50:50 AS:GH, and 25:75

AS:GH. Eating and rumination behaviors were recorded every 5 minutes for 24 hours.

Rumination time per unit of DMI decreased with decreasing dietary particle size (Table

2-1). The shortest rumination time was 4.6 hours/day or 13.5 minutes/kg of DMI. Total

chewing time varied from 12.1 to 9.9 hours/day, and decreased with decreasing MPL.

Ruminal pH was measured every 5 seconds for 48 hours by an industrial electrode placed

in the ventral sac via ruminal cannulae. Ruminal pH was evaluated according to time

above 6.2 and below 5.8, based on the observations of Russell and Wilson (1996) that

ruminal microbial activity is compromised when pH falls below 6.2, and the premise that

the incidence of sub-clinical acidosis increases when ruminal pH drops below 5.8. Cows

fed 25:75 AS:AH had a greater mean ruminal pH (P=0.10) and time above pH 6.2

(P=0.08) than cows fed 50:50 AS:AH, perhaps due to decreased DMI (P<0.10) and

increased time spent eating (P=0.01). Ruminal pH status was mainly affected by forage

particle size, and was improved by feeding chopped, rather than ground, alfalfa hay

(Table 2-1).

Ruminal pH is influenced by the relative concentrations of acids, bases and buffers

present at any given time. Fermentation ofNFC found in high concentrations in

concentrates and cereal grains results in rapid production of organic acids, while peNDF

consumption stimulates saliva flow. Concurrent consumption ofNFC and peNDF









sources can be facilitated by combining diet components in a TMR rather than offering

forage and concentrate separately. Maekawa et al. (2002) used eight ruminally

cannulated lactating Holstein cows arranged in a double 4 x 4 Latin square design to

examine the effects of offering whole crop barley silage and steam-rolled, ground barley

grain-based concentrate on chewing activities, saliva production and ruminal pH. Four

diets were offered: separate ingredients (SI) offered at a forage to concentrate (F:C) ratio

of 50:50 (DM basis), and adlibitum TMR with F:C ratios of 60:40, 50:50, and 40:60.

When SI were fed, silage was offered ad libitum and grain was offered at 50% of

previous DM consumption to facilitate consumption of a 50:50 F:C ratio. However,

animals fed SI elected to consume an actual F:C ratio of 43:57, illustrating that when diet

components are fed separately, the animals may choose to consume a different F:C ratio

than the formulated ration. Cows were fed twice daily, with silage fed 1 hour after the

concentrate on SI treatment. Eating and ruminating activities were recorded every 5

minutes for 24 hours. Ruminal pH was measured every 5 seconds for 24 hours, and mean

pH for each 15-minute period was recorded. Saliva samples were collected at the cardia

during feeding and resting. Salivation could not be measured during rumination, so

salivation rate was assumed to be the same as that observed during eating.

Salivary secretion rate was 2.2 times greater during eating than resting. Linear

effects of increasing silage concentrations in the TMR showed only numerical increase in

minutes/ day spent eating (P=0.18), but increased minutes/ day spent ruminating

(P=0.03), increasing total chewing time (P=0.01) from 741 to 757 and 848 minutes/day

on F:C ratios of 40:60, 50:50, and 60:40, respectively. Total chewing time by cows fed

the SI diet (736 minutes/ day) was similar to that of cows fed the 40:60 TMR. However,









SI was consumed at a faster (P=0.02) rate (10.9 minutes/ kg of DM) than the 40:60 TMR

(13.5 minutes/ kg of DM). Concentrates required less mastication before being

swallowed, and were therefore consumed more rapidly than silage or TMR (P<0.01),

decreasing the amount of saliva produced per unit of feed (ml/g of DM) from 4.43 on

silage to 1.19 on concentrates (P<0.01). The main diet effect on ruminal pH was not

significant, but numerical differences presented in Table 2-2 suggest that diet did alter pH

characteristics. The postprandial decline in ruminal pH was greatest for cows fed SI, and

mean ruminal pH in SI cows was below the 5.8 benchmark suggested for increased risk

of subacute ruminal acidosis. Feeding TMR rather than SI appeared to decrease risk of

ruminal acidosis by preventing higher than intended consumption of concentrates and

increasing saliva production via increased mastication at the time that rapidly fermentable

concentrates were consumed.

Effects of Particle Size on Digesta Passage, Intake, and Fermentation

Kennedy et al. (1992) reported a linear increase in DMI (P<0.001) and linear

decreases in ruminal DM retention time (P=0.001) and DM digestibility (P<0.01) when

wheaten hay fed ad libitum to sheep and goats was progressively switched from hay

chopped to 1 cm (c), to a 2:1 ratio of c + hay ground to pass through a 3-mm screen and

pelleted (p), to a 1:2 ratio of c+p, to an all p diet. Dietary particle size can influence

voluntary intake, rate of digesta passage, and rate and extent of ruminal fermentation.

Since the rumen is a dynamic system, multiple interactions, exceptions, and caveats must

be considered. The following includes a brief, and by no means comprehensive,

discussion of such considerations.

Although multiple factors, including hydration and density, influence the rate at

which feed particles leave the rumen, resistance to outflow increases with increasing









particle size (Poppi et al., 1980). Poppi et al. (1980) suggested a high resistance for

particles greater than 1.18 mm, since less than 5% of particles leaving the rumen of sheep

fed chopped (2 cm) hay were retained on a 1.18 mm mesh sieve. Small particles may

flow from the rumen more rapidly than large particles (Weston and Cantle, 1984). Long

forage particles form a mat in the rumen, which can trap small particles, retaining them

for a greater length of time (Grant, 1997). As a result, decreasing forage particle size

may decrease overall ruminal DM retention (Bernard et al., 2000).

An increased rate of digesta passage can increase voluntary intake by reducing the

constraints of rumen fill. A direct relationship has been demonstrated between increased

ruminal contents and decreased voluntary intake (Campling and Balch, 1961); Schettini

et al., 1999). When long forages are consumed, rumen distention resulting from

restricted digesta flow may limit voluntary intake (Allen, 1996), which may prevent

intake of sufficient energy to meet animal requirements (Miller and O'Dell, 1969).

Conversely, fine particle size in an entire ration, particularly in high concentrate rations,

can also reduce voluntary intake (Krause and Combs, 2003), likely due to unfavorable

alterations in rumen conditions, such as a decline in pH (Nocek, 1997).

Dietary particle size impacts ruminal digestibility in part by influencing the balance

between the length of time feed components are retained in the rumen and the rate at

which they are fermented. Ruminal fermentation of dietary nutrients takes time. An

increased rate of particulate outflow from the rumen decreases the amount of time feed

components are available for microbial fermentation. Ruminal digestibility of dietary

components, particularly of slowly fermenting fiber, may decrease as a result (Pasha et

al., 1994). Conversely, chopping or grinding feed components increases the surface area









available for microbial access (Owens and Goetsch, 1988), increasing the rate of

fermentation. As a result, decreasing the particle size of a particular feed component may

increase the extent of its digestion in the rumen (Callison et al., 2001). However, when

rapid fermentation is not balanced by increased buffers and disappearance of organic

acids from the rumen, ruminal pH will decrease, the microbial population will be altered,

and nutrient digestibility may decrease (Russell and Wilson, 1996)

Changes in dietary physical form alter the ruminal environment, and may result in a

shift in VFA profile. In a study by Krause et al. (2002), four TMR rations with forage to

concentrate ratios of 39:61 (DM basis) were used to investigate the effects of forage

particle size and the concentration of dietary ruminally fermentable carbohydrates (RFC)

in diets of equal NDF content. Diets were offered ad libitum to eight ruminally

cannulated, lactating Holstein cows. Dietary concentrates were cracked-shelled corn

(low RFC) or ground high-moisture shelled corn (high RFC). Dietary forages were

chopped alfalfa silage of mean particle length of 13.6 (coarse) or 3.7mm (fine).

Decreasing forage particle size decreased the amount of time engaged in chewing

behavior (Table 2-3). Both decreasing forage particle size and increasing RFC decreased

mean ruminal pH, increased time (hours per day) and area (time*pH units/day) of

ruminal fluid below pH 5.8, and decreased the ratio of acetate to propionate by increasing

ruminal propionate concentration (Table 2-3). The percentage of dietary particles

retained on the top screen of a Penn State particle separator was positively correlated

(0.61) with minutes of chewing per day (P 0.0003) and, to a lesser extent, negatively

correlated (-0.32) with time that ruminal fluid pH was <5.8 (P 0.09).









Jorgenson and Schultz (1963) fed lactating cattle 7.26 kg of ground corn daily,

along with long-stem (control) or pelleted alfalfa hay in ad libitum amounts. Feeding

pelleted hay increased DMI from 1.1 to 1.2 kg/ 45 kg of body weight (P<0.05), increased

total ruminal VFA concentration from 600 to 844 mg/ 100 ml of ruminal fluid (P<0.05),

and altered the VFA profile of ruminal fluid. As a percentage of total VFA, acetate

decreased from 60.3 to 55.6% (P<0.05), propionate increased from 20.1 to 27.0%

(P<0.05), and butyrate decreased numerically from 17.0 to 15.3% (P>0.05). A second

trial compared the same control diet with an experimental diet of 16.3 kg of a 50:50

mixture of alfalfa: corn pellets, plus ad libitum alfalfa pellets. Changes in VFA profile of

the ruminal fluid were similar to the first trial. Intake, however, declined from 1.22 to

1.18 kg/ 45 kg of body weight (P<0.05) due to difficulty in keeping the cows consuming

the pelleted feed. Blood glucose concentrations increased from 49.1 to 57.4 mg%

(P<0.05) in cows fed the all-pellet diet. Although ruminal pH was not reported in this

study, subacute ruminal acidosis (SARA) may be caused by an elevation in total VFA

(Stone, 2004), and can result in decreased or variable intake (Nocek, 1997).

In diets for domestic ruminants, forage has long been the staple physically effective

component, and the effects of forage particle length on digestion parameters and

production performance have been investigated widely. In more recent years, byproduct

(non-forage) feeds have come into use as effective fiber sources in the livestock industry.

The peNDF value of nonforage fiber sources is considerably lower than long-stem

forages, but may be higher than some forms of concentrates, grains, and ground forages

(Mertens, 1997).









Partial replacement of forage with cottonseed hulls (CSH) has been shown to

increase DMI and decrease NDF and DM digestibility (Moore et al., 1990; Theurer et al.,

1999) in cattle, suggesting an associated increase in passage rate. Dry matter intake as a

percent of body weight in steers fed sorghum grain and chopped (15 cm screen) alfalfa

hay increased (P<0.05) when half of the alfalfa was replaced with CSH or chopped (2.5

cm screen) wheat straw (Theurer et al., 1999). However, DMI intake per kg of body

weight gain was lower for steers fed alfalfa, suggesting an increased efficiency of feed

utilization with alfalfa as compared to CSH.

Further effects were noted when CSH replaced long (theoretical 22.3 mm) or short

(theoretical 4.8 mm) cut corn silage in TMR offered to lactating Holstein cows (Kononoff

and Heinrichs, 2003). The inclusion of CSH at 8% of dietary DM reduced the

concentration of corn silage from 57.4 to 45.8% of dietary DM. Physically effective

NDF, estimated according to (Mertens, 1997), did not differ across treatments, and

increased sorting behavior was noted when cows were fed long corn silage. It is

therefore not surprising that reducing corn silage particle size did not affect (P>0.05)

chewing activity, DMI, ruminal pH, or apparent total tract digestibility of total

nonstructural carbohydrates, NDF or acid detergent fiber (ADF). Total chewing time per

kilogram of NDF intake tended (P<0.10) to increase by cow fed diets containing long

corn silage. The inclusion of CSH tended (P=0.06) to decrease total chewing time per

kilogram of DM consumed, and decreased mean ruminal pH from 6.24 to <6.17

(P=0.05). Cottonseed hulls increased DM and NDF intake (P<0.01) but did not affect

DM digestibility.









Potential Implications of peNDF for Captive Giraffe

From 1974 to 1978, a series of digestion trials comparing grass and alfalfa hays

were conducted using thirty herbivore species at five zoos (Foose, 1982). For each

dietary treatment, 14 days were allowed for acclimation to diets, followed by 10 days of

intake measurements. Total fecal production was collected during the final 4 days.

Individual animal intake was recorded, and orts were composite by species and

treatment (grass or alfalfa hay) for chemical analysis. Daily fecal output of each animal

was weighed, and pooled subsamples from each of the 4 days were composite for

analysis. The report mentions that "in most cases, it was possible to collect (fecal)

accumulations only at 24 hour intervals." Given that inability to collect samples soon

after defecation resulted in sample trampling and contamination that prevented total fecal

collection in one giraffe study (Clauss et al., 2001), giraffe digestibility results from the

Foose (1982) study may be inaccurate.

Intake and nutrient extraction results are reported for three giraffe fed alfalfa hay

and one giraffe fed timothy grass hay (Table 2-4). Foose (1982) reports that grass hay

digestion trials on several giraffe had to be discontinued because the animals were

"conspicuously languishing on the diets." Giraffe used were 2 and 3 years old, and thus

were likely to still be growing. Body weight could not be measured, and was estimated

from the literature. Organic matter intake (% of BW) and digestion coefficient ((amount

ingested amount defecated) / amount ingested) x 100% were 0.45 and 57.11 for grass

hay, and 0.89 and 60.70 for alfalfa hay. Of the 28 ungulate species in this study, giraffe

had the lowest intake as a percentage of body weight, suggesting that the ruminal fill

effects of long forage may have an especially high impact in giraffe.









Feeding all-forage diets to high producing dairy cows can prevent adequate energy

intake, but feeding high-concentrate rations has potential to cause a variety of metabolic

and health problems (Miller and O'Dell, 1969). It has been suggested (Clauss et al.,

2002b) that captive giraffe fed a traditional hay/ concentrate diet face a nutritional

dilemma. Giraffe consuming a high proportion of hay will increase ruminal fill and

decrease intake. Those that consume a high proportion of concentrates will increase their

potential risk of ruminal acidosis.

The appropriate particle size for maintaining optimal diet utilization and

gastrointestinal health in CS remains to be determined, but it has been suggested that

giraffe are ill-suited to consumption of long-stem forages and the formation of a ruminal

mat similar to that in domestic ruminants (Clauss et al., 2002b). The giraffe's diet in the

wild consists primarily of polygonal leaves as opposed to the elongated grasses and hays

consumed by GR (Clauss et al., 2002b). These physical as well as chemical features

including density factors may explain why stratification of ruminal contents occurs in

wild GR, but not in wild CS (Hofmann, 1973; Clauss et al., 2001). Anatomical

differences between GR and CS (Hofmann, 1973) suggest a naturally faster passage rate

of digesta in CS ruminants than in GR ruminants, which has been observed in captive

giraffe (Hatt et al., 1998). Giraffe appear to be adapted to rapid rates of fermentation and

nutrient absorption (Hofmann, 1973).

However, feeding high-concentrate, low-forage diets is unlikely to enhance captive

giraffe nutrition. It seems likely that even in giraffe, acid production from ruminally-

degraded organic matter (particularly rapidly fermenting NFC) must be balanced with the

ruminal dilution and salivary stimulation effects of an appropriate form of peNDF (Allen,









1997). Saliva production in wild giraffe may be high due to the oral stimulation and time

involved in selective feeding patterns. One of the challenges in feeding captive giraffe is

providing diets with a physical form that will stimulate mastication and saliva flow to

maintain a balanced rumen pH, while avoiding an unnatural reduction in rate of passage

and dietary intake.

Dietary Carbohydrate Profile

Carbohydrate Fractions

Dietary carbohydrates can be divided into two basic fractions: fiber and nonfiber

carbohydrates (NFC). The chemical bonds between sugar residues in dietary fiber cannot

be broken by mammalian enzymes. As a result, fiber cannot be digested by mammals

themselves, but may be utilized by microbes in the digestive tract. The end products of

microbial fermentation can then be absorbed and utilized by the host animal.

Fermentation of cellulose and hemicelluloses is typically slow (2 to 14% digestion/ hour)

(Sniffen et al., 1992). Nonfiber carbohydrates, on the other hand, are generally

associated with rapid rates of ruminal fermentation, with digestion rate constants of 75 to

400% digestion/ hour for sugars, and 5 to 50% digestion/ hour for starch and pectin

(Sniffen et al., 1992). With the exception of neutral detergent-soluble fiber (NDSF),

NFCs may be digested by mammalian enzymes. The NFC may be further divided into

sugars (mono-, di-, and oligosaccharides), starch, organic acids, and NDSF (Hall et al.,

1999). As we will see, mounting evidence suggests that these NFC fractions differ in

their fermentation characteristics.

As the science of carbohydrate analysis has progressed, new techniques have

allowed evaluation of carbohydrates in increasing detail. Use of different techniques

over time can make the literature difficult to interpret. For the sake of clarity,









carbohydrate fractions found in the different analytical systems discussed in this section

are listed below.

Proximate analysis system

Crude fiber (CF): cellulose, acid- and alkali-insoluble hemicellulose, acid- and

alkali-insoluble lignin

Nitrogen-free extract (NFE): sugars, starch, pectic substances, organic acids,

fructans, acid- and alkali-soluble hemicellulose, acid- and alkali-soluble lignin

Under the proximate analysis system, fiber (hemicellulose) and lignin can appear in

either the CF or NFE fraction. The detergent system, however, distinguishes between the

fiber and non-fiber carbohydrate fraction of feeds.

Detergent system

Neutral detergent fiber (NDF): cellulose, hemicellulose, lignin

Acid detergent fiber (ADF): cellulose, lignin

Nonfiber carbohydrates (NFC): sugars, starch, pectic substances, organic acids,

fructans, and other carbohydrates soluble in neutral detergent

Carbohydrates in Natural CS Diets

The methodology for differentiation of NFC into sugars, starch and NDSF has only

recently been applied to animal feedstuffs. (Dierenfeld et al., 2002) examined sugar and

starch concentrations in 8 fruit and 2 flower species consumed by wild duikers in the

Democratic Republic of Congo. Samples contained 2 tol5 times more sugar than starch,

with DM percentages of sugars at 0.16 in one sample and 3.19 to 15.71 in 9 samples, and

DM starch percentages of 7.43 in 1 sample, 0.37 to 1.96 in four samples, and less than

0.1 in 5 samples. Duikers are considered frugivores (fruit eaters), consuming large









proportions of fruits, flowers and seeds. Hence, no leaf material was included in this

report.

Unlike duikers, giraffe are considered folivores (foliage eaters). Plants consumed

by wild giraffe vary widely with season and geographic location. While Acacia spp.

appear to be the most commonly reported dietary item ( Foster, 1966; Dagg and Foster,

1976; Furstenburg and Van Hoven, 1994; Ciofolo and LePendu, 2002; Caister et al.,

2003), giraffe in a given location may consume as many as 66 plant species over the

course of a year (Leuthold and Leuthold, 1972). Trees and shrubs comprise the bulk of

the diet, with limited vine and herb consumption (Leuthold and Leuthold, 1972). Grass

consumption appears either non-existent (Leuthold and Leuthold, 1972) or negligible

(Ciofolo and LePendu, 2002). Field and Ross (1976) observed that woody plants

comprised > 93% of the diet, and "grass appeared to be eaten by accident when enmeshed

with other food." Plant portions consumed are primarily leaves and stems, but may also

include fruits, flowers and bark (Leuthold and Leuthold, 1972; Caister et al., 2003). The

diet of giraffe in Niger was reportedly composed of 86% leaves, 8.5% stems and 5.5%

flowers and fruits. However, during the dry season (December to April), the diet was

composed of 45% leaves and stems, 44% fruits and 11% flowers (Ciofolo and LePendu,

2002).

Wild CS are generally thought to consume a "rich, rapidly fermenting diet"

(Hofmann, 1973). Pellew (1984) reported mean daily nutrient intakes of 2.22 to 3.12 kg

of crude protein (CP), 0.53 to 0.97 kg of ether extract (EE), 4.33 to 8.63 kg of acid

detergent fiber (ADF), and 5.48 to 7.73 kg of nitrogen-free extract (NFE) for giraffe in

the Serengeti. Composition of leaves (DM basis) from 10 plant species consumed by









giraffe in Niger during the dry season ranged from 8.2 to 28.6% CP, 0.8 to 6.8% crude

fat, 1.55 to 16.2% crude fiber (CF), and 19.8 to 71.93% NFE (Caister et al., 2003). Mean

composition of browse leaves (DM basis) from the Narus Valley (Uganda) for the

months of January, March, April, May and December ranged from 11.51 to 22.37% CP,

2.36 to 3.01% EE, 19.33 to 33.48% CF, and 36.80 to 50.32% NFE. Browse stems

contained 5.78 to 9.35% CP, 1.17 to 1.62% EE, 32.49 to 48.80% CF, and 35.91 to

52.18% NFE (Field and Ross, 1976). The high NFE content in these studies seems to

suggest a high concentration of nonfiber carbohydrates. However, it is important to

remember that NFE contains not only sugars, starch and soluble fiber, but also variable

amounts of hemicellulose and lignin, and therefore is not a true representation of NFC

content.

In our search of the literature, we found no published data on NFC fractions in

foliage. Analysis of leaves collected in October from three tree, one shrub, and one grass

species used for zoo animal enrichment showed higher concentrations (DM basis) of

sugars (6.5 to 13.2) than starch (0.3 to 3.1) in four of five species (Table 2-5; R. Ball,

personal communications). Acacia leaves contained 37.2% NFC, 6.5% sugars, and 0.5%

starch, suggesting that NDSF and organic acids may account for approximately 81% of

the NFC fraction. As will be discussed below, the differing NFC fractions do not digest

in the same manner in a fermentative environment. In order to improve our

understanding of their nutritional requirements, and optimize nutrition of captive

folivorous CS species, NFC analysis of natural dietary foliage needs to be further

explored.









Effects of NFC Source on Fermentation Characteristics and Animal Performance

Domestic ruminant feeding studies suggest that high- and low- starch NFC sources

differ in digestion characteristics and fermentation end products. Grains such as corn,

oats, wheat, and barley contain starch as the main nutrient component (Huntington,

1997). Supplemental energy sources having less starch include molasses and sucrose as

sugar sources, as well as beet pulp and citrus pulp, which have substantial NDSF and

sugar contents.

Carbohydrate feeding trials that have involved addition or subtraction of whole

feeds generally alter multiple ration characteristics, rather than carbohydrate profile

alone. This may explain the variability in DMI and body weight changes in livestock fed

diets in which NDSF or sugar sources replaced fibrous or starchy feeds. Increasing

supplemental NDSF sources (generally beet pulp or citrus pulp) while decreasing starchy

feeds (generally barley or corn) has both increased DMI in cattle (Chester-Jones et al.,

1991) and had no effect on DMI in sheep (Ben-Ghedalia et al., 1989). Supplemental

NDSF tended (P = 0.05) to decrease average daily gain in one study (Chester-Jones et

al., 1991), but showed no effect on body weight in others (Bhattacharya and Lubbadah,

1971; Friggens et al., 1995).

Altering dietary NFC profile has changed fermentation patterns, ruminal pH, and

diet digestibility. Switching lactating cows fed a basal diet of freshly cut grass (85%

Lolium perenne) from a supplement (36% of dietary DM) containing 47.5% corn meal

and 50% hominy to a supplement containing 82.5% sugar beet pulp and 15% soybean

hulls increased (P<0.05) ruminal OM and NDF digestibility and VFA concentrations

(Van Vuuren et al., 1993). Ben-Ghedalia et al. (1989) used four ruminally and

duodenally cannulated Merino rams to compare digestion characteristics of supplements









of dried citrus pulp (84.4% of supplement DM) with supplement containing citrus pulp

(20.4%) and barley (76.5%). Isonitrogenous supplements DCP (84.4% citrus pulp) and B

(20.4% citrus pulp, 76.5% barley) were offered with lucerne hay in an 80:20 supplement:

forage ratio. Sheep consuming DCP had greater ruminal pH (6.42) than those consuming

B (6.18). Total tract apparent digestibility of OM did not differ between diets, but

apparent NDF digestibility was greater by sheep fed DCP (79.4%) than those fed B

(63.6%). The results indicate that the fermentation of DCP created a more favorable

ruminal environment for fermentation of NDF.

In vitro investigation of microbial CP yield, detected as trichloroacetic acid-

precipitated crude protein (TCACP), also suggested differences in NFC fermentation

patterns (Hall and Herejk, 2001). When isolated NDF from bermudagrass hay was

incubated with sucrose, corn starch, or citrus pectin in media containing mixed ruminal

microbes, maximum TCACP synthesis (mg/g substrate OM) was greater for starch (85.6)

than sucrose (73.3) or pectin (75.4) (P <0.05). Temporal patterns of TCACP yield

differed as well. Sucrose showed no detectable lag phase, with greatest yield of TCACP

at the 12-hour sampling, and declined only gradually over time. The TCACP reached

peak yield at 12 and 16 hours during fermentation of substrate containing pectin,

followed by a rapid decline. When starch was fermented, peak TCACP yield was not

reached until 16 hours, also followed by a rapid decline.

Bhattacharya and Lubbadah (1971) used cattle and sheep to examine the effects of

replacing corn with beet pulp in high (approximately 75%) concentrate diets. Three

experiments were conducted, with the same four supplements fed in each experiment. In

all three experiments, the control supplement (Diet I) consisted of corn (73%), soybean









meal (21%), tallow (0.8%), bonemeal (2.7%), limestone (2.0%) and salt (0.5%). Beet

pulp was used to replace 0, 50, 75, and 100% of the corn in Diets I, II, III, and IV,

respectively. As beet pulp concentrations increased, concentrations of tallow and

bonemeal were increased, and limestone decreased, to maintain similar levels of fat,

phosphorus, and calcium among diets.

Experiment I used four lactating Holstein-Friesian cows in a 4 x 4 Latin square

design, with 4-week periods, to evaluate dietary effects on body weight gain, milk yield,

and milk fat. Animals were offered 19 kg of concentrate and 5 kg of alfalfa hay daily.

No significant differences were observed. It was noted that bloat symptoms occasionally

seen during pre-experimental and experimental periods in cows on the control diet were

not observed in cows offered the beet pulp diets.

Experiment II consisted of two 17-day trials using eight wethers housed in

metabolism crates, with two animals randomly allotted to each feeding treatment during

each trial. Alfalfa hay in this experiment was ground (size unreported) and mixed with

concentrate in a 26.2: 73.8 ratio. Diets were offered twice daily for three hours each

feeding. Sample collection in each trial included total feces and urine (final 7 days) and

jugular blood samples at three hours post-feeding on the day following the end of the

collection period. Apparent total tract DM digestibility was not affected by diet.

Replacement of Diet I (73% corn supplement) with Diet IV (73% beet pulp supplement)

decreased the apparent percent digestibility of CP (78.2 to 72.5%) and NFE (87.1 to

85.5%) and increased the apparent percent digestibility of CF (46.1 to 72.4%) (P < 0.01).

No dietary effect on blood glucose or VFA concentration was reported.









In Experiment III, rations "similar to those in Experiment 2" were offered ad

libitum to four ruminally fistulated Holstein steers. Beginning with the control ration and

progressing to higher beet pulp concentrations, animals received each ration for seven

days, with four days gradual adjustment between diets. Intake (average 7.65 kg/ day) did

not differ across diets. Ruminal fluid measures at the end of each period showed

differences (P<0.01) in VFA concentration (mmole/ L) response among all treatments,

and increased VFA concentrations in animals on Diets II (138.7), III (157.5) and IV

(141.3) as compared to control animals (113.3). Ruminal pH was numerically lower on

Diet IV (5.9) than on Diets I, II, and III (6.2, 6.3, and 6.2 respectively), and lactic acid

concentration was not affected. The authors observed that when animals received the

control diet, "the rumen was full of gas and frothy foam. As the proportion of ... beet

pulp increased..., gas and foam gradually disappeared, and the appearance of the rumen

ingesta resembled that under hay-feeding conditions."

Notable treatment differences in fermentation characteristics were observed in

ruminally cannulated Angus x Hereford steers (n=20) offered low-quality tallgrass-prairie

hay (chopped through a 75-mm screen, offered at 130% of previous intake) only

(control), and hay supplemented with starch, glucose, fructose or sucrose at 0.30% of

BW/ day (Heldt et al., 1999). Supplemented steers also received ruminally degradable

protein (RDP) (sodium caseinate, 91.6% CP) at 0.031% and 0.122% of BW/ day in

Experiments 1 and 2, respectively. Supplements were administered intraruminally. In

Experiment 2, total tract OM digestibility was 7.9% greater in animals supplemented with

sugars as compared to those receiving starch (P =0.04). In agreement with Ben-Ghedalia

et al. (1989), ruminal pH and total tract NDF digestibility were lower for animals









supplemented with starch than for those supplemented with sugar (P <0.05). Volatile

fatty acid concentrations differed between control and supplemented animals, and

between animals supplemented with sugar or starch (Table 2-6). Ruminal concentrations

of butyrate and lactate were greater in animals supplemented with sugars than in control

animals or animals supplemented with starch (P <0.01). In both experiments, time-

related fermentation patterns were dramatically different for sugars and starch. With all

three sugar treatments, pH declined rapidly, reaching the lowest point at 3 hours after

supplementation, followed by a rapid recovery. Decline in pH was slower for the starch

treatment, reaching the lowest point 9 hours after supplementation.

Interaction Between Dietary Components

Heldt et al. (1999) also illustrates the possible impact of other dietary components,

in this case protein, on fermentation characteristics of NFC fractions. In Experiment 1,

where RDP was fed at 0.031% of BW/ day, OM and NDF digestibility did not differ

between supplement treatments, and NFC supplementation depressed NDF digestibility

when compared to the hay-only control. Increasing RDP from 0.031 to 0.122% BW/ day

(Experiment 2) resulted in higher OM and NDF digestibilities in animals receiving

supplement vs. control, and sugars vs. starch. Carbohydrate treatment affected mean

ruminal pH at high, but not low, RDP amounts. Other factors beyond RDP can alter the

effect of dietary NFC. The physical form of the NFC source can affect the site of

digestion and the extent of ruminal digestion (Callison et al., 2001). The proportion of

dietary forage can affect ruminal VFA production response to individual NFC feeds

(Friggens et al., 1998). Dietary characteristics other than NFC profile alone may affect

animal response to NFC supplementation.









Potential Implications of Dietary NFC Profile for Captive Giraffe

Clemens et al. (1983) reported the molar proportions of VFA in the digestive tracts

of 16 species (4 CS, 5 intermediate feeders (IM), and 7 GR) of free-ranging African

ruminants, using animals sacrificed during wildlife management programs. The digestive

tract was separated by ligatures into six segments, and representative content samples

from each segment were strained through cheesecloth. Supernatant was acidified with

concentrated H2SO4 and refrigerated for later analysis. Sample collection and field

analysis were generally completed within 1 hour after death. Of all species, giraffe had

the highest ruminal proportion of acetate (73.2 + 1.6 molar %; 60.2 to 72.9 molar % in

other species), and the second lowest proportion of propionate (14.1 + 0.5 molar %; 12.8

to 22.8 molar % in other species) in ruminal fluid. The acetate: propionate ratio (5.21 +

0.13) in ruminal fluid from giraffe was exceeded only by the eland (5.71 + 1.38) and

followed closely by the oryx (5.13 + 0.11), with other species ranging from 2.92 to 4.65.

Traditionally, high acetate concentration is associated with fiber fermentation, and

elevated propionate concentration is associated with NFC fermentation. Since giraffe are

thought to consume diets relatively rich in NFC and low in NDF, with extensive fiber

fermentation impeded by a rapid rate of digesta passage from the rumen, high acetate/

low propionate concentrations in the wild giraffe rumen appear paradoxical. However,

domestic ruminant studies reveal that the various NFC fractions differ in fermentation

and VFA production characteristics. Although VFA disappearance rates and the

unknown nutrient content of feedstuffs consumed prior to sample collection cannot be

discounted as possible causes for the findings of Clemens et al. (1983), another potential

contributing factor lies in the still unquantified NFC fractions of giraffe diets consumed

in the wild.









In domestic sheep and cattle, ruminal acetate and propionate profiles have been

altered by modification of dietary NFC. When high amounts of citrus pulp (84.4% of

dietary DM) were fed to sheep in place of a 20.4% citrus pulp and 76.5% barley diet,

propionate decreased numerically from 17.6 to 14.4 molar % (P>0.05) and acetate

increased from 65.0 to 69.1 molar % (P<0.05) (Ben-Ghedalia et al., 1989). Using

continuous culture in vitro fermentations with mixed ruminal microbes from cattle

inoculum, Ariza et al. (2001) observed changes in molar concentrations of propionate and

acetate when starch was decreased from 24.0 to 11.0% and NDSF was increased from 8.8

to 14.4% by altering the substrate ratio of hominy feed to citrus pulp. Propionate

decreased from 22.7 to 16.7 molar % and acetate increased from 62.6 to 68.9 molar %

(P<0.04). The acetate: propionate ratio increased from 2.8 to 4.1 (P=0.01). Strobel and

Russell (1986) provided six carbohydrate substrates (starch, sucrose, cellobiose, xylan,

pectin, and a mix of the preceding in equal parts) to mixed ruminal bacteria at 1 mM/ h

for 10 h at an initial pH of 6.7 (neutral) or 6.0 (low). Propionate concentrations across

treatments were numerically lowest for pectin, though the differences were not significant

(P>0.05) at low pH. Less (P<0.05) propionate was produced in response to pectin (1.3

mM) than starch (2.9 mM) or cellobiose (2.7 mM) at pH 6.7. At neutral pH, millimolar

concentrations of acetate from fermentation of pectin, cellobiose, starch, mixed

carbohydrates, sucrose, and xylan were 10.1, 6.4, 5.1, 4.8, 4.7, and 3.6 respectively, and

were greatest (P<0.05) from pectin fermentation. At low pH, pectin fermentation

resulted in greater (P<0.05) acetate concentrations than fermentation of other single-

carbohydrate treatments, and numerically greater (P>0.05) acetate than from mixed

carbohydrates. Acetate concentrations from pectin fermentation at low pH were half (5.0









mM) those observed at neutral pH (10.1). It should also be noted that decreasing pH

increased (P<0.05) lactate production from starch (0.9 to 4.1 mM), sucrose (3.7 to 8.3

mM), and cellobiose (2.3 to 8.2 mM) whereas lactate was not detectable from

fermentation of pectin or xylan at either pH. The NFC profile of giraffe diets consumed

in the wild has not been determined, but high concentrations of pectic substances in

native feeds could explain the ruminal acetate and propionate concentrations observed in

free-ranging giraffe.

The extraordinarily high ruminal papillation in wild giraffe results in an average

24x increase in ruminal surface area (Hofmann, 1973), denoted as surface enlargement

factor (SEF; (papillary surface + base surface)/ base surface, where papillary surface =

length x mid-level width of papillae x 2 (Hofmann et al., 1988)). By contrast, mean

ruminal SEF in two captive giraffe was 1.90 and 2.67, owing to a dramatic decrease in

papillae size and density (Hofmann and Matern, 1988). The diet of the two captive

giraffe was ad libitum lucerne hay, browse (amount not reported), variable amounts of

produce, and a concentrate mixture consisting primarily of oats, barley, germinated

wheat, soya, lucerne pellets, and maize.

Butyrate is thought to be the most influential VFA in ruminal papillae development

(Van Soest, 1994). Changes in ruminal butyrate concentrations have been variable when

sugar or pectin sources were fermented in combination with varying proportions of starch

or peNDF. It is interesting to note that in both NFC experiments conducted by Heldt et

al. (1999), ruminal concentrations of butyrate were lower than propionate for animals

consuming control and starch-containing treatments, but greater in animals consuming

diets supplemented with sugar, due to an increase in butyrate concentrations. Strobel and










Russell (1986) also observed greater (P<0.05) in vitro butyrate production from

fermentation of sucrose as compared to starch at neutral (6.7) pH, but no difference at

low (6.0) pH. These observations raise questions about a potential relationship between

high ruminal SEF in wild giraffe and VFA production from fermentation of the unknown

NFC fractions of foliage compared with low SEF in captive giraffe offered high-starch

concentrates.



Table 2-1. Effects of particle size of alfalfa-based dairy cow diets on chewing activity
and ruminal pH.
---------------Dietsa.--


AS:AH 50:50


AS:AH 25:75


Item CH GH CH GH SE PC
MPLb, mm 9.78 5.13 7.50 4.42 0.56 0.01
Eating activity (min/kg DM) 12.5 10.8 14.8 15.5 1.7 >0.15
Rumination time (min/kg DM) 19.1 15.7 20.8 13.5 1.6 0.01
Total chewing time (min/kg DM) 31.6 26.5 35.6 28.9 2.7 0.01
Mean rumen pH 5.97 5.78 6.18 5.90 0.15 0.02
pH > 6.20, hours 7.3 3.5 11.0 7.6 3.3 0.11
pH < 5.80, hours 7.5 13.0 5.1 11.7 2.6 0.01
pH > 6.20, area (pH*hours) 6.9 10.5 4.4 10.2 2.2 0.01
pH < 5.80, area (pH*hours) 2.1 3.7 0.9 4.5 1.1 0.01


aAS:AH = Ratio of alfalfa silage to alfalfa hay; CH = chopped hay; GH
barley-based concentrate and 40% forage on a DM basis.


ground hay; overall diet was 60%


bMPL = Mean particle length. Fifty percent of dietary particles are greater or less than this length.
P-value for forage particle size effect.
(Beauchemin et al., 2003)

Table 2-2. Effects of concentrate level and feeding management on ruminal pH of
lactating dairy cows.
------------------Diets-----------------
SIa ---Total Mixed Ration---
Item 43:57b 40:60b 50:50b 60:40b SE Pc
Mean pH 5.77 5.84 5.99 5.98 0.08 0.25
Minimum pH 5.14 5.26 5.40 5.30 0.06 0.06
Percent of time under pH 5.8 55.36 46.25 33.25 34.78 8.0 0.26
Area under pH 5.8 (pH*hours/ day) 4.99 3.52 2.60 2.37 0.81 0.19
aSeparate ingredient diet.
bForage-to-concentrate ratio consumed.
P-value for main effect of diet. Linear and quadratic contrasts calculated for the proportion of silage in the
TMR were not significant (P>0.05).
(Maekawa et al., 2002)











Table 2-3. Effects of forage particle size and grain fermentability on chewing activity,
ruminal pH, and ruminal VFA profile in midlactation cows.
Particle


---------Chewing--------


Ruminal VFA


--Ruminal pH--


Dieta
HMCFS
HMCCS
DCFS
DCCS
SED
Forageg
RFCg


Mean mm
3.0
6.0
2.8
6.3


Eatb
10.1
9.8
9.8
12.2
0.8
0.09
0.07


Rumc
63
96
54
83
6
0.0001
0.03


TCTd
24.9
30.6
21.7
30.9
1.7
0.0001
0.24


mM
161.5
148.4
151.1
144.9
5.6
0.03
0.10


A:Pe
1.60
1.90
2.23
2.45
0.12
0.03
0.0001


Mean
5.72
5.98
5.90
6.07
0.08
0.0006
0.02


h<5.8f
14.3
7.2
9.3
5.5
1.4
0.0001
0.003


Forage x
RFCg 0.02 0.60 0.16 0.40 0.68 0.39 0.11
aHMCFS = High-moisture corn and fine silage, HMCCS = high-moisture corn and coarse silage, DCFS =
dry corn and fine silage, DCCS = dry corn and coarse silage, SED = standard error of the difference.
bMinutes per kilogram dry matter intake
"Minutes of rumination per kilogram NDF intake
dTotal chewing time, minutes per kilogram dry matter intake
eRuminal acetate to propionate ratio
fHours per day below pH
9P-value for effect of variable
(Krause et al., 2002; Krause and Combs, 2003)

Table 2-4. Nutrient intake and digestion coefficients from giraffe fed all-hay diets.
----- ------ Intake--------------------- Digestion coefficient
Treatment OM OM CW CP OM CW CP
Units % BW kg/kg BW0.75 kg/kg BW075 g/kg BW0.75 % % %
Grass 0.45 0.025 0.019 1.20 57.11 57.51 86.01
Alfalfa 0.89 0.049 0.026 10.90 60.70 55.92 93.76


Alfalfa SE 0.01 0.006 0.006 1.40
Values for 1 and 3 giraffe on ad libitum grass and alfalfa hay diets, respectively.
OM= Organic matter; CW = cell wall; CP = crude protein
Modified from Foose (1982).


6.19 2.19


size


1.50










Table 2-5. Chemical composition (DM basis) of five browse plants grown at Busch
Gardens in Tampa, Florida.


Item
(DM)
Ash
NDF
NFC
Sugar
Starch
CP
NDICP
Ca
P


Hibiscusa
18.7
16.87
53.9
15.8
3.2
3.9
15.4
8.4
3.31
0.77


Bambooa
47.7
11.6
65.1
12
6.9
0.3
12.9
5.6
1.08
0.28


NDF = Neutral detergent fiber; NFC
protein.
aLeaf
bLeaf + stem


Acaciaa
39.7
14.23
33.7
37.2
6.5
0.5
16.5
5.2
4.23
0.24


Acaciab
46.8
9.58
46.1
34.8
4.6
0.8
13.6
7.2
2.99
0.24


Mulberrya
27
11.48
33.6
34.6
13.2
3.1
25.7
9.4
2.16
0.4


non-fiber carbohydrates; NDICP


False
Acaciaa
28


7.88
36.4
43.3
7.1
0.7
18.3
9.1
1.9
0.22


False
Acaciab
52.2
6.05
56.7
28.9
4.4
2.1
10.2
4.3
1.63
0.17


neutral detergent insoluble crude


Samples collected in October, 2004.
Unpublished data courtesy of Busch Gardens, Tampa.










Table 2-6. Influence of supplemental carbohydrate source fed in combination with 0.122% BW/ day of degradable intake protein on
ruminal fermentation characteristics.
-------------Orthogonal contrasts, P-value-------------
Control Starch Mono- Glucose
vs vs vs vs
Component Control Starch Glucose Fructose Sucrose SEM supplement sugar disaccharide fuctose
pH 6.56 6.13 6.16 6.29 6.22 .04 <.01 .04 .97 .03
NH3 N, mM .31 2.42 2.99 2.85 1.88 .47 <.01 .79 .10 .83
OA, mA4 70.8 98.9 96.0 89.0 89.1 2.91 <.01 .05 .37 .12
Acetate 73.5 69.5 61.5 59.8 59.7 .71 <.01 <.01 .34 .11
Propionate 14.0 16.4 14.1 14.2 14.4 .31 .05 <.01 .40 .81
Butyrate 10.6 10.3 17.5 18.9 19.5 .61 <.01 <.01 .11 .13
Isobutyrate .63 .82 .72 .66 .64 .05 .14 .02 .42 .39
Valerate .46 1.43 1.74 1.73 1.72 .06 <.01 <.01 .30 .92
Isovalerate .60 1.13 .99 .94 .90 .08 <.01 .07 .54 .71
Lactate .31 .45 3.52 3.82 2.95 .58 <.01 <.01 .33 .72
aOA = Total organic acids (VFA + lactate)
VFA concentrations presented as mol/100 mol
(Heldt et al., 1999)














CHAPTER 3
EFFECTS OF ALTERING THE PHYSICAL FORM AND CARBOHYDRATE
PROFILE OF THE DIET ON CAPTIVE GIRAFFE

Introduction

Physical characteristics of feeds, particularly of the fiber fraction, can influence the

efficiency of use of a ruminant's diet. Proper ruminal function in domestic ruminants is

maintained by consumption of sufficient "physically effective fiber" (peNDF), which is

determined by the neutral detergent fiber (NDF) concentration and the ability of the feed

to promote a chewing response in the animal (Mertens, 1997). The appropriate physical

form, digestibility, and quantity of peNDF for concentrate selectors (CS) such as giraffe

have not been determined. It has been proposed that peNDF for CS includes particles of

polygonal shape similar to browse, rather than the needle-like fibers derived from grasses

(Clauss et al., 2002). Captive giraffe fed a traditional hay/ concentrate diet may face a

nutritional dilemma: consume a high proportion of hay, increasing ruminal fill and

decreasing intake, or consume a high proportion of concentrates, increasing potential risk

of ruminal acidosis (Clauss et al., 2002).

Nonfiber carbohydrates (NFCs) can be divided into sugars (mono-, di-, and

oligosaccharides), starch, organic acids, and neutral-detergent soluble fiber (NDSF),

which includes carbohydrates such as pectin (Hall et al., 1999). The NFC fractions differ

in fermentation characteristics and end products. Sugar (sucrose), starch and pectin have

been shown to differ in vitro in both maximum microbial protein yields and in temporal

pattern of that yield (Hall and Herejk, 2001). When fermented in the absence of other









NFC sources, sugars have produced higher butyrate concentrations than starch (Heldt et

al., 1999; Strobel and Russell, 1986). Strobel and Russell (1986) noted that unlike

fermentation of sugars or starch, pectin fermentation did not result in detectable lactate

production at either neutral (6.7) or low (6.0) ph media. Fermentation of pectin or feeds

high in NDSF has increased acetate concentrations both in vitro (Ariza et al., 2001;

Strobel and Russell, 1986) and in vivo (Ben-Ghedalia et al., 1989).

Although the NFC fractions of giraffe diets in nature have not been quantified

extensively, analysis of leaves collected from five tree and shrub species used for

enrichment of zoo animal diets showed greater DM concentrations of sugars (6.5 13.2)

than starch (0.3 3.1) in four of five species (R. Ball, personal communications). Acacia

leaves contained 37.2% NFC, 6.5% sugars, and 0.5% starch, suggesting that NDSF,

organic acids, and analytical may account for approximately 81% of the NFC fraction (R.

Ball, personal communications). Consumption of feeds high in NDSF by wild giraffe

could explain why when (Clemens et al., 1983) examined molar proportions of volatile

fatty acids (VFA) in the digestive tracts of 16 free-ranging African ruminant species, the

highest ruminal proportion of acetate and the second highest acetate: propionate ratio

were found in giraffe. Collectively, these observations raise questions about the

importance of dietary NFC complement in maintaining optimal ruminal fermentation,

and the response of captive giraffe to the various NFC fractions.

Nutrient supply to the ruminant animal is not dependent upon diet composition

alone. The amounts and proportions of offered dietary items that the animal chooses to

consume, digestion characteristics and interactions of consumed dietary components, and

nutrient absorption also must be considered. It has been suggested that the wasting and









sudden death widely reported in captive giraffe (Fox, 1938; Chaffe, 1968; Fowler, 1978;

Strandberg et al., 1984; Flach et al., 1997; Junge and Bradley, 1993; Ball et al., 2002)

may be a result of a functional energy deficiency ( Fowler, 1978; Ball et al., 2002). The

reports of ruminal acidosis (Clauss, 1998; Clauss et al., 2002b), fermentative gastritis or

rumenitis (Fox, 1938; Ball et al., 2002), gastrointestinal ulceration (Fox, 1938; Fowler,

1978) and pancreatic pathologies ( Fox, 1938; Fowler, 1978; Lechowski et al., 1991; Ball

et al., 2002) also may be indicative of consumption of an unbalanced diet or altered

fermentation of dietary components in captive giraffe. Compounding this problem,

decreased ruminal absorptive surface area that has been reported in captive giraffe

(Hofmann and Matern, 1988) may impair nutrient absorption.

The objective of this study was to examine the effects of modified dietary physical

form and NFC profile compared to a commercial diet on captive giraffe intake, nutrient

digestibility, behavior, and blood parameters. The hypothesis was that modifying

supplement physical form and NFC profile would alter the intake, nutrient digestion,

behavior, and blood parameters of captive giraffe fed a hay + supplement diet.

Materials and Methods

Design

This study was conducted from August 2002 through February 2003 at Busch

Gardens in Tampa, FL (BGT). The design was a modified reversal study using two

treatments and six animals in seven 21-day periods (Table 3-1). The study was

conducted according to protocols approved by the Institutional Animal Care and Use

Committees of the University of Florida and Busch Entertainment Corporation.









Giraffe

Six non-lactating adult female reticulated giraffe (Giraffa camelopardalis

reticulata) were used. Body weight (BW) ranged from 576 to 715 kg (average 637 +

42.6 kg). Giraffe were assigned identification numbers G1 through G6, in order of

entrance to the study. Three giraffe (G2, G4 and G6) were wild-caught, and estimated to

be approximately 22 years of age at the time of the study. Three giraffe were born at

BGT and aged 12 (G1), 7 (G3), and 4 (G5) years at time of entrance to the study.

Pregnancy status of the animals varied. Two giraffe were in the third (G1) and

second (G3) trimesters of pregnancy during their time on study, and the remaining four

giraffe were non-pregnant. Nine and a half months prior to entrance to the study, G5

delivered a large (74 kg), rapidly growing calf. The calf was weaned through progressive

separation two to five days prior to G5's entrance to the study.

Facilities

Study giraffe were individually housed in two adjacent pens, which allowed

continual access to visual and tactile contact with conspecifics. Pen dimensions were

approximately 22 and 21 meters (east to west) by 8 and 9 meters (north to south) in pens

1 and 2 respectively. Flooring in each pen consisted of a roughed concrete pad of

approximately 9 by 9 meters spanning the west end, and sand in the remainder of the pen.

A shade shelter sloped east to west (9 and 7 meters high respectively) and covered the

western 1/3 of each pen. Radiant heaters attached to the bottom of the shade shelter

provided heat when temperatures fell below 100C. Pen 1 was equipped with a second

shelter adjacent to the west end of the pen that provided feed troughs with additional

protection from driving rain. Feed in pen 1 suffered less weather damage than feed in

pen 2 for one collection day in period 1 as a result. Water was provided in a 189-litre tub









at ground height in pen 1, and an automatic waterer (Nelson Manufacturing Company,

Cedar Rapids, IA) at approximately 1.8 meters high in pen 2.

Hayracks with catch pans and supplement feed tubs were mounted to the center of

the west wall in each pen. Feeders were only accessible to the animal in that pen.

Hayracks and supplement feeders were placed adjacent to one another, with the top of

each mounted to the top of the wall at approximately 3.2 meters high. At the onset of

period 3, the supplement feed pan in pen 2 was lowered 15 cm to accommodate the

shorter height of G3, and remained at this height for the remainder of the study.

Diets

The standard BGT giraffe ration consisted of adlibitum alfalfa hay, fresh browse as

available, and two concentrates, Mazuri Browser Breeder and Purina Omelene 200

(Purina Mills, St. Louis, MO). Concentrates were fed together in a 75:25 ratio (as-fed

basis) at approximately 1% of giraffe body weight as per manufacturer's

recommendations for Mazuri Browser Breeder. This 75:25 grain mix (GF) was used as

the control diet for comparative evaluation to the coarse, non-pelleted experimental

browser feed (EF). Ingredient composition (DM basis) of EF was 35% sugar beet pulp,

18% cottonseed hulls, 13% molasses, 11% soybean meal, 10% alfalfa meal, 6% mineral

mix*, 4% heat-treated soybean meal (Soy Plus) and 3% sucrose (table sugar).

*Mineral mix was 94.5% DM, and was composed of the following ingredients
(percentage inclusion on air dry basis): dicalcium phosphate (39.5%), calcium carbonate
(8.76%), urea (8.76%), salt (5.25%), sodium bicarbonate (17.5%), potassium carbonate
(9.65%), selenium 1% (0.12%), cobalt sulfate (0.02%), copper sulfate (0.08%), zinc
sulfate (0.08%), iron sulfate (0.30%), manganese sulfate (4.38%), magnesium oxide
(4.38%), ethylenediamine dihydroiodide (0.003%), vitamin A 650 x 106 IU/ kg (0.03%),
vitamin D3 400 x 106 IU/ kg (0.01), and vitamin E 500 x 103 IU/ kg (0.80%). Mix
contained: 84% ash, 24% CP as non-protein N, 9.9% Ca, 8.3% P, 6.9% Na, 3.2% Cl,
2.6% Mg, 4.7% K, 0.96% S, 68 ppm Co, 203 ppm Cu, 20 ppm I, 5270 ppm Fe, 14118
ppm Mn, 12 ppm Se, 2060 ppm Zn, 88452 IU/# Vit. A, 18144 IU/# Vit. D3, 1814 IU/#
Vit. E.











The EF was formulated to be equivalent to GF in minerals and vitamins A, D, and E, and

differed from GF and other commonly-fed browser feeds by containing less starch, more

sugars and soluble fiber, and small, heavily lignified particles (cottonseed hulls) to

modify the fiber size and texture of the diet. The experimental supplement was mixed in

178 kg batches by study personnel at the University of Florida Dairy Research Unit

(Hague, FL).

Each animal was housed individually and fed EF or GF ad libitum for 21 days, and

then received the other feed supplement in the subsequent 21-day period, so that each

animal received each diet. Alfalfa hay, water, and salt were offered ad libitum

throughout the study.

Sample Collection and Analyses

Feedstuffs and intake

Amounts of all feedstuffs offered and refused were weighed and recorded daily.

Subsamples of all feedstuffs were collected. In periods 1 through 3, two to three flakes of

alfalfa hay from a bale fed (and considered representative of alfalfa offered during that

period) were retained for analysis. In periods 4 through 7, a bale corer was used to

sample five to seven of the bales fed during that period. On days 15-21 of each study

period, th total amount of hay and supplement orts for each giraffe were collected and

frozen at -230C until analysis.

Dry matter (DM) content of supplement fed and daily supplement orts were

obtained by drying duplicate subsamples of two to three grams each at 1050C in a forced

air oven for approximately 36 hours, until a constant weight was achieved. Results were









used to calculate daily dry matter intake (DMI) for each animal. Subsamples of offered

supplements and supplement orts were dried at 550C in a forced-air oven until constant

weight was achieved. Subsamples were retained for particle size analysis, and the

remaining dried sample was ground through a Wiley mill (A.H. Thomas, Philadelphia,

PA) to pass through a 1-mm screen.

In order to obtain representative subsamples for DM determination, hay offered in

periods 1 through 3 and total daily hay orts for each animal were chopped with a paper

cutter to roughly 2.5 cm, then blended and subsampled following the procedure of Van

Soest and Robertson (1985). One subsample of approximately 10 grams was used to

determine DM. Remaining sample was dried in a forced-air oven at 550C, ground

through a Hammer mill (Smalley Manufacturing Co., Manitowoc, WI) using a 0.635-cm

screen, mixed and subsampled. Subsamples were ground through a Wiley mill using a 1-

mm screen. Ground samples were analyzed for DM at 1050C and organic matter (OM)

by ashing overnight (>8 hours) at 5120C in a muffle furnace (Sybron Corporation,

Dubuque, IA). Daily orts were composite on a DM basis by giraffe by period. Batches

of EF fed during the collection week of each period were composite by period. Mazuri

Browser Breeder and Purina Omelene 200 fed in each period were analyzed separately

for nutrient content, and results used to calculate the nutrient composition of GF (75%

Mazuri Browser Breeder, 25% Purina Omelene 200) fed in each period.

Feed offered and composite orts were analyzed for DM, OM and nutrient content.

Neutral detergent fiber (NDF) and NDF organic matter (NDFOM) were determined using

heat-stable ca-amylase (Goering and Van Soest, 1970; Van Soest, 1991). Acid detergent

fiber (ADF) (Goering and Van Soest, 1970), sugar (Hall, 2001), starch (Hall, 2001),









lignin (Goering and Van Soest, 1970), and mineral (using Perkin Elmer 3300 XL ICP

manufactured by Perkin Elmer, Shelton, CT) (Association of Official Analytical

Chemists, 1990) content were analyzed by Cumberland Valley Analytical Services,

Maugansville, MD. Neutral detergent soluble fiber was determined according to the

method of (Hall et al., 1999). Crude protein (CP) as N x 6.25 was determined by

combustion analysis (Association of Official Analytical Chemists, 1990) using Macro

Elementar Analyzer (vario MAX CN, Elementar Analysensysteme GmbH, Hanau,

Germany). Nutrient intake was calculated as: kg nutrient offered kg nutrient refused.

Particle size of supplements offered was analyzed using U.S.A. Standard Testing

Sieves (Fisher Scientific Company, Pittsburg, PA) 5, 10, 18, 23, 60, and 120 (pore sizes

of 4, 2, 1, 0.5, 0.25, and 0.125 mm). Samples of approximately 10 g DM were soaked in

300 ml of deionized water for 2 hours. Hydrated samples were quantitatively transferred

to stacked sieves. The top sieve was rinsed for 5 minutes using a three-jet water sprayer

calibrated to a flow of 1 L per minute (D. Mertens, personal communication), then

inverted and contents rinsed into a Gooch crucible under vacuum. This procedure was

repeated for each sieve. Crucibles were dried at 1050C for approximately 36 hours to

determine DM retained on each screen. Results were used to calculate modulus of finess

(MOF) according to Poppi et al. (1980).

Fecal collection and analysis

Attempts at total fecal collection began on day 16 in period 3, and day 15 of each

remaining period, and lasted for 56 108 consecutive hours. Duration of total fecal

collection in a given period was often limited by adverse weather conditions or animal

behavior that resulted in sample loss or safety risk to the animal or staff. In period 3,









attempts to collect the total daily fecal production of G3 were considered unsuccessful

due to excessive pacing by that animal. Total collection was not attempted on G3 in

period 4. Samples were collected in gallon Ziploc bags approximately six times per day,

as staff became available, and for two hours overnight on days 16 and 17 in period 3, and

days 15 and 16 in remaining periods. Samples were frozen at -230C for subsequent

analysis.

Fecal samples from each giraffe were composite by each 24 hours of collection,

yielding two consecutive 24-hour composites for each giraffe in each study period, with

the exception of G3 in P4, when total collection was not attempted. Compositing

procedure was as follows. Samples were thawed in their original bags overnight at room

temperature. Each individual bag of fecal material was weighed on a Mettler PE 3600

top-loading balance. Bag contents were emptied into a plastic container. All non-fecal

organic debris and as much contaminating sand as possible were removed from the

sample and returned to the original bag. Bag and debris weight was subtracted from

initial weight to calculate sample wet weight. Sample was mixed manually, and

subsamples of 10% + 0.02% of original wet weight were taken from each bag for each

composite. Two composites were retained for each giraffe by day, and frozen at -130C

until analysis.

Composite 1 was dried for analysis of DM, OM, NDF and CP. The thawed sample

bag was laid flat and rolled with a plastic cylinder to crush fecal pellets and blend

material, facilitating accuracy of subsampling. Duplicate subsamples of approximately 2-

3 g each were used to determine DM and OM and the remaining sample was dried in a

forced air-oven at 550C and ground to pass through the 1-mm screen of a Wiley mill.









Crude protein as N x 6.25 was determined in duplicate on 0.25 0.35 g samples by

combustion analysis (Association of Official Analytical Chemists, 1990) (Vario MAX

CN, Elementar Analysensysteme GmbH, Hanau, Germany). Duplicate samples of 0.7 g

each were used to determine NDFOM using heat-stable a-amylase (Goering and Van

Soest, 1970; Van Soest, 1991 ). Percent digestibility of NDFOM and apparent percent

digestibility of OM and CP were calculated as: (mean g nutrient consumed per day) -

(mean g nutrient defecated per day) / (mean g nutrient consumed per day), using intake

values from days 14 through 20 in each period and fecal samples collected for 48

consecutive hours on days 16 through 18 in periods 3 and 6, and days 15 through 17 in

remaining periods. Nutrient kg digested was calculated as: nutrient kg consumed x

percent digestibility.

Composite 2 was crushed manually, subsampled for DM and OM, and analyzed

wet for CP, to examine for volatile nitrogen (N) loss due to drying and grinding.

Composite 1 and composite 2 samples were analyzed for CP using macro-Kjeldahl

analysis (Association of Official Analytical Chemists, 1990).

Behavior

On days 13 through 15 of each period, behavior was recorded for 48 consecutive

hours by on-site trained observers. Rumination, consumption (supplement, hay, water,

salt and sand), oral stereotypes (tongue play, licking metal and licking non-metal

inanimate objects), tactile contact between giraffe, and locomotor behaviors (standing,

walking, lying down with head erect and lying down with head on flank) were recorded

every sixty seconds using instantaneous sampling (Martin and Bateson, 1986). Results

were analyzed for differences in the number of minutes over 48 hours engaged in each

individual behavior, total oral stereotypes, supplement + hay consumption, supplement +









hay consumption + rumination, and total oral behavior (all oral stereotypes + total feed,

water, salt and sand consumption + rumination).

Body weight and blood samples

A 2.4 meter high solid wooden chute system adjacent to pen 1 provided manual

restraint for blood collection. The chute was equipped with a scale for measuring body

weight. At approximately 0800 hours on day 1 of the first period an animal entered the

study and day 21 of each period, weight and body condition scores (scored by C. Kearney

on a scale of 1 through 8) were recorded and blood samples collected on each animal

before feeding. Blood samples were collected via jugular venipuncture using manual

restraint (chute system). Blood samples were collected on all giraffe using clot tubes and

three anti-coagulants: ethylenediaminetetraacetic acid (EDTA), sodium heparin (NaH),

and sodium citrate (NaC) (Vacutainer Brand tubes by Becton Dickinson, Franklin Lakes,

NJ). Blood was collected using an 18- or 14-gauge needle, 30-inch extension set and 35-

cc syringe. Blood samples were transferred into the tubes, placed on ice, and returned to

the zoo hospital laboratory within two hours of collection. Sodium citrate samples were

placed on a rocker until shipping for fibrinogen analysis.

Complete blood count (on whole blood prepared with EDTA) (Automated Cell

Counter) and serum chemistry profiles (Hitachi 747-200 Chemistry Analyzer, Hitachi,

Hitachi-NAKA, Japan) were analyzed by Antech Diagnostics (Largo, FL). Serum

chemistry profiles evaluated glucose, urea nitrogen, creatinine, total protein, albumin,

total bilirubin, alkaline phosphatase, alanine aminotransferase, aspartate

aminotransferase, cholesterol, Ca, P, Na, K, Cl, Mg, globulin, lipase, amylase,

triglycerides, creatinine phosphokinase, gamma-glutamyltransferase, lactate

dehydrogenase, and calculated osmolality. Additional plasma and serum were pipetted









into scintillation vials and frozen at -800C for later analysis. Plasma with NaH was

analyzed for insulin (double antibody radioimmunoassay procedure; Soeldner and

Sloane, 1965), and non-esterified fatty acids (NEFA) (NEFA-C kit; Wako Fine Chemical

Industries USA, Inc., Dallas, TX; as modified by Johnson and Peters (1993) at the

University of Florida.

Statistical analysis

Data were analyzed by the MIXED procedure of SAS (1999). Animal response

data were analyzed using a statistical model that included animal, period, and diet with

animal as a random variable. Supplement nutrient content and fecal nutrient content were

analyzed using the model "nutrient = diet". Results are reported as least squares means

with standard errors as determine with PROC MIXED. Pearson correlation coefficients

were calculated using the PROC CORR procedure of SAS. Due to the small number of

animals in this study (n=6), significance was declared at P<0.10, and tendency at

0.10
Results and Discussion

As planned, the EF diet contained greater concentrations of sugar (P<0.001) and

NDSF (P<0.001), and decreased concentrations of starch (P<0.001) (Table 3-2).

Concentrations of ash, lignin, and NDFOM were also greater in EF, and CP

concentrations tended to increase as well, and variable differences occurred in mineral

concentrations (Table 3-2). Supplement particle size differed (P<0.001) between diets.

The EF supplement was courser than GF, with a greater MOF value (4.97) than GF

(3.15).









Intake

Average daily DM intake varied greatly among the six animals for both alfalfa hay

(0.12 to 3.94 kg/day) and supplement (2.87 to 9.26 kg/day) consumption, but did not

differ between diets (Table 3-3). As giraffe shifted from consuming supplement GF to

EF, starch intake decreased (P=0.052) from 0.93 to 0.12 kg/ day, sugar intake tended

(P=0.115) to increase from 1.12 to 1.53 kg/day, and NDSF intake increased (P=0.074)

from 0.85 to 1.19 kg/day. Consumption of ADF (1.83 vs. 2.23 kg/day) (P=0.039) and

lignin (0.33 vs. 0.50 kg/day) (P=0.064) differed between animals consuming supplement

GF and those consuming EF, respectively.

Although supplements were formulated to be equivalent in concentration of

minerals, CP, and NDF, analysis of the supplements fed revealed slight differences in

these analytes (Table 3-2). As a result, treatment-associated differences for mineral

intake (Table 3-3) may reflect differences in supplement mineral concentrations.

Differences in supplement nutrient concentration and intake may account for the

increased consumption of NDFOM and ash when giraffes were offered EF. There was a

tendency for greater CP consumption by giraffe offered EF than those offered GF

(P=0.136).

Pellew (1983) estimated intake of wild giraffe in the Serengeti by recording the

number of bites during feeding bouts, then multiplying by mean bite mass. Mean bite

mass was first estimated by hand clipping foliage to simulate giraffe browsing behavior,

then corrected using observations of the number of bites taken by captive giraffe

consuming a known quantity of browse during timed feeding periods. Daily DM intake

was estimated as 19.0 kg for males and 16.6 kg for females. Using average live weights

of 1200 kg and 800 kg, as reported by Dagg and Foster (1976), mean daily DM intake









was estimated at 1.6 and 2.1% of BW for males and females, respectively. Dry matter

intake as a percentage of BW in the present study ranged from 0.69 to 1.66% and

averaged 1.22% (+ 0.28), much lower than intake reported for wild giraffe (Pellew,

1983), but similar to DM intake by non-lactating captive giraffe reported by Baer et al.

(1985) of (1.22% of BW) and by Clauss et al. (2001) of (0.97 to 1.28% of BW).

Digestibility

Several challenges occurred with attempted total fecal collection. In period 1, total

collection was attempted on days 15 and 16, but was considered suspect on day 16

because of possible sample loss due to heavy rain. When laboratory analyses yielded

unusually high digestibility values for day 16, it was concluded that total collection likely

had not been achieved, and this day was excluded from data analysis. Digestibility data

from G3 in period 3 and G4 in period 4 were also excluded from data analysis, since

collection records, high sample ash content, and high digestibility results suggested that

excessive trampling had prevented total collection of fecal material and given excessive

contamination with sand. Since collection was not attempted on G3 in period 4, no

digestibility data is reported for period 4.

Because samples were collected from sand-bedded pens, sand contamination was a

continual complicating factor. Although samples were "cleaned" manually at the time of

collection and again prior to compositing, ash content of fecal composites ranged from 20

to 63% of sample DM (20 to 46% in samples included in results). Sand contamination

increased subsampling error, which often increased variation between duplicates when

samples were analyzed for OM, CP, and especially NDFOM at the University of Florida.

Subsampling of fecal samples and analysis of NDFOM was evaluated at the U.S. Dairy

Forage Research Center (USDA-ARS, Madison, WI), where the same difficulties were









encountered. In the end, any samples not meeting the Horwitz criteria (Horwitz, 1982)

for difference between duplicates were re-analyzed in duplicate at the University of

Florida, and the mean of four values was reported.

Previous attempts at determining apparent digestibility of DM, NDF, and CP in

captive giraffe fed hay/concentrate diets using acid detergent lignin, acid insoluble ash,

indigestible NDF and alkanes as markers have been summarized by Clauss et al. (2001).

Mean apparent digestibility of CP, OM, and NDFOM in the present study fell within the

range of apparent digestibility of CP (59.7 to 82.4%), DM (52.0 to 85.2%), and NDF

(32.5 to 74.8%) previously reported for captive giraffe (Clauss et al., 2001). Digestibility

of NDFOM averaged 55.5 + 4.0%, and was not affected by treatment (Table 3-3). The

kg of NDFOM digested increased (P=0.077) by 0.25 kg on EF, and was likely due in part

to the 0.47 kg increase in NDFOM intake on EF. Apparent OM digestibility averaged

72.0 + 4.5%, and decreased numerically (P=0.206) on the EF diet. Apparent CP

digestibility was numerically 10 percentage units greater for GF than EF (P=0.151), but

apparent kg of CP digested differed by only 0.06 kg (Table 3-3). The EF supplement

included 4% heat-treated soybean meal, containing protein which is less degradable in

the rumen. We do not know if a ruminally undegradable protein source was present in

GF. This potential difference between supplements may have contributed differences in

true CP digestibility between diets. Alternatively, an increase in yield of ruminal or

hindgut microbes may have facilitated an increased presence of microbial CP in the fecal

material of giraffe consuming EF.

The CP concentrations of feces from giraffe fed EF were greater (P<0.004) than

concentrations for giraffe fed GF using all three analytical techniques (Table 3-4). When









fecal samples were analyzed for CP in dry and wet forms, CP concentrations were

generally higher in the wet samples by approximately two percentage units, suggesting a

loss of volatile nitrogen during the drying process. Apparent CP digestibility calculated

using dried feces may have been overestimated as a result.

Behavior

The effect of dietary treatment on giraffe behavior is reported in Table 3-5. The

number of minutes engaged in rumination and hay consumption over 48 hours was not

affected by treatment. Time engaged in supplement consumption was 2.3 times longer

by giraffe offered EF than by giraffe offered GF (P=0.063), and total feeding time tended

(P=0.100) to be greater by giraffe offered EF. The tendency (P=0.124) for increased

time spent in eating and rumination by giraffe fed the EF diet has potential to have

increased saliva flow and ruminal buffering.

Leuthold and Leuthold (1972) reported on the diurnal time budgets of wild giraffe.

Female giraffe spent 53.1% and 15% of daylight hours engaged in feeding and

rumination, respectively (Table 3-6). By contrast, giraffe in this study spent 17.7% and

22% of time feeding and ruminating. It is interesting to note that oral stereotypes, which

have not been observed in wild giraffe, increased the total time captive animals spent

engaged in oral behavior from 39.7% for feeding and rumination alone to 54.1%, which

bears more similarity to the 68.1% of time wild giraffe spent engaged in feeding and

rumination.

Oral stereotypy appears to be the most prevalent stereotypic behavior observed in

captive giraffe. A survey of giraffe and / or okapi-holding American Zoo and Aquariums

Association (AZA) accredited institutions by Bashaw et al. (2001) yielded data on the

occurrence of stereotypes in 214 giraffe and 29 okapi in 49 institutions. The most









prevalent stereotypic behaviors were repetitive licking of non-food objects (referred to as

"licking") and pacing. Licking was reported in 72.4% of giraffe + okapi, and pacing was

reported in 29.9%. Additional behaviors reported in 3.2% of animals included head

tossing, self-injury, and tongue playing. An attempt to reduce incidence of stereotypic

behavior had been made by 51.7% of responding institutions, with reported success in

51.9% of these institutions. Tarou et al. (2003) attempted using Bitter Apple, a

chemical spray used to deter undesirable chewing in horses, to deter stereotypic licking in

three captive giraffe. Animals simply shifted licking behavior to a non-treated area.

Bashaw et al. (2001) suggested that licking in captive giraffe and okapi may be

related to feeding motivation. In the present study, oral stereotypes were recorded as

three separate oral behaviors: repetitive licking of metal objects, repetitive licking of non-

metal objects, and tongue play unassociated with feeding, rumination, drinking, or

licking. Licking metal was the most prevalent oral stereotype (mean = 258 minutes / 48

hours), followed by tongue play (mean = 105 minutes / 48 hours) and licking non-metal

objects (mean = 20 minutes / 48 hours). The amount of time spent engaged in tongue

play tended (P=0.119) to be twice as long in giraffe offered GF (Table 3-5). Although all

six animals exhibited each of the individual stereotypes, they varied in individual

preference for metal licking or tongue play (Figure 3-1). The number of minutes over 48

hours spent engaged in total oral stereotype behavior ranged from 209 to 661, with a

mean of 383 + 127. Despite the increase in time engaged in feeding behavior, minutes

engaged in total oral stereotypes decreased only numerically (P=0.223) from 433 to 318

minutes in giraffe offered EF.









During this study, giraffe were repeatedly observed drooling or swallowing while

engaged in oral stereotype behavior. Although this observation is purely subjective in

nature, it suggests that oral stereotypy may also facilitate some amount of ruminal

buffering.

Blood Measures

Few blood parameters were affected by diet (Table 3-7). Animals consuming GF

had higher (P=0.028) blood glucose values (99.0 mg/dl), compared to those consuming

EF that showed values (82.3 mg/dl) more similar to the range reported for domesticated

ruminants (40-80 mg/dl) (Swenson, 1984). Switching captive giraffe from GF to EF

decreased starch consumption and increased NDSF consumption. Ruminal fermentation

ofNDSF rather than starch has increased acetate and decreased propionate concentrations

in ruminal fluid in domestic livestock studies. Since acetate may be utilized directly as

an energy source or used as a lipogenic substrate, whereas propionate is converted largely

to glucose, a potential NFC profile-associated shift in ruminal VFA production offers one

possible explanation for the decreased blood glucose concentrations in giraffe fed EF.

Despite the similarity in apparent kg of CP digested, blood urea nitrogen showed a

numerical decrease (P=0.166) from 20.6 mg/dl in giraffe fed GF to 16.6 mg/dl in those

fed EF. Comparatively for animals consuming EF, this may reflect a shift in dietary

protein utilization towards proliferation of microbial mass, decreased degradation of body

protein, or decreased degradation of absorbed amino acids.

Non-esterified fatty acids (NEFA) decreased numerically for animals consuming

EF (P=0.282), and decreased on EF in all giraffe except G1 (Figure 3-2), who elected to

consume a 99% supplement diet when fed EF. A BW gain of> 15 kg occurred in 5 of 6

giraffe when EF was fed, and only in 1 of 6 giraffe when GF was fed, but treatment-









associated changes in BW varied among animals and were not significant (P = 0.327)

(Table 3-3). These observations raise the question of the basis for possibly decreased

adipose tissue mobilization with diet EF. Could part of this effect be the result of

increased NDSF intake and increased mass of NDFOM digested altering ruminal

acetate:propionate ratio, providing the animals with more lipogenic nutrients?

Blood glucose was positively correlated with kg of OM and CP digested and

consumed (Table 3-8). Although BUN showed a positive correlation with apparent kg of

CP digested, a stronger positive correlation existed between BUN and starch intake.

Glucose and BUN were positively correlated with one another, and of the blood proteins

measured, CPK was the only protein positively correlated with both glucose and BUN

(Table 3-9). Increased blood concentrations of CPK have been associated with increased

muscle catabolism.

Ancillary Study Observations / Individual Animal Effects

The giraffe originally designated to serve as G4 became ill at the time of entrance

to the study. As a result, the study was halted for three weeks, and this animal eventually

entered the study as G6. During the break in the study, G3 continued to receive the EF

diet, which she had received in the previous period. Over the 6 weeks on EF, her body

weight increased by 35 kg, from 542 to 577 kg. While pregnancy status may have been a

contributing factor, her weight gain during three weeks on GF was only 3 kg. This

observation, coupled with increased body weight and condition in a lactating non-study

giraffe fed EF for several months suggest that greater changes in animal response may

have been observed had study periods been of longer duration. Furthermore, ancillary

observations suggest that feeding EF to two lactating giraffe not on study may have

increased milk production. When these females were switched from the normal giraffe









ration (GF) to EF, the fecal consistency of their calves was loose the following day, and

gradually returned to normal over the subsequent days.

By using six non-lactating adult female giraffe, this study contained the largest and

most similar giraffe population reported in a captive feeding study. Even so, each animal

proved to be a unique individual, exhibiting qualities or behaviors that no doubt affected

study results. Giraffe 1 elected to consume a diet of 98.7% supplement and 1.3% hay

when fed the EF diet. Giraffe 2, an arthritic older female with a historically strong

exhibition of maternal behavior, was on the study for three periods, and received the EF

supplement in both periods 1 and 3. The low mean voluntary intake for G2 in period 1

(0.69% of BW), which was lower than other reported intake values (0.86 to 1.66% of

BW), may have been related to factors other than dietary treatment. During the eleven

acclimation days in period 1 during which as-fed intake values can be considered

reasonably accurate (days without heavy rain), her daily as-fed intake averaged 7.01 kg,

similar to mean DM intake (6.92 kg) during the collection week of period 3. However,

mean daily intake on days 16-18 of period 1 decreased by nearly half, to 3.58 kg (3.31 kg

of DM). This may have been related to environmental factors rather than diet. Sand

erosion caused by heavy rain had made it necessary for G2 to step up onto the concrete

pad in order to access feeders. While attempting to walk onto the concrete pad on day

18, G2 was observed "tripping" over a small irrigation pipe that had been exposed by

sand erosion. After the pipe was removed during the morning feeding on day 19, G2

would not step up onto the concrete, and consumed no feed that day. Consequently, day

19 was removed from analysis of intake results. During morning feeding on day 20, a

sand ramp was constructed, and feeding behavior resumed. However, G2 did not









approach the feeders after the evening feeding on day 20, appearing instead to be

preoccupied with the birth and overnight keeper observations of a giraffe calf in an

adjacent pen. Thus, intake amounts on day 20 (3.15 kg of DM) were similar to intake on

days 16 through 18.

Giraffe 3 had not been housed in a small pen prior to this study, and had a history

of nervous behavior around humans. She spent a great deal of time pacing, displaying

nervous behavior which may have affected intake and blood values, and certainly

prevented total fecal collection. Giraffes 3 and 4 frequently engaged in tongue play with

the mouth closed, rather than open, making the behavior difficult to detect from a

distance. Consequently, oral stereotype behavior may have been underestimated or

disproportionately reported in these animals.

Giraffe 5 was on the study for three periods, and received GF supplement in

periods 5 and 7. Prior to study entrance, this primiparous female had been nursing a

rapidly growing calf, and had shown a decline in body weight and condition. Body

weight increased steadily from 611 kg at study entrance to 661 kg at study exit, and body

condition score increased from 3.5 to 4.5 on a scale of 1-8, suggesting that G5 was

recovering from the demands of lactation during her time on the study. Giraffe 5 also

elected to consume hay as a substantially greater proportion (39 to 41%) of total DM

intake than other giraffe. In short, individual animal variation may have played a large

role in the results of this study.

Diet Selection

Both dietary physical form and carbohydrate profile play a role in development or

prevention of ruminal acidosis. Voluntary intake patterns of low amounts of forage and

high amounts of concentrate by captive giraffe appear similar to those associated with









ruminal acidosis in domestic livestock. Hay: supplement intake ratios for giraffe in the

current study ranged from 1:99 to 41:59 and averaged 21:79 (DM basis). Previously

reported hay: concentrate intake ratio of two giraffe fed in the same pen was 26:74 (Baer

et al., 1985). Another study using four giraffe supplemented with limited amounts of

browse in addition to lucerne hay and concentrates found forage:concentrate intake ratios

of 25:75 to 51:49 (Clauss et al., 2001). The forage: supplement intake ratios of the 21

measures taken from these three studies averaged 27:73 (DM basis). Decreasing dietary

ratios of forage: concentrate from 40:60 to 30:70 has been used to induce subclinical

ruminal acidosis in dairy cattle (Krajcarski-Hunt et al., 2002). Although giraffe are

believed to be adapted to diets characterized by a rapid ruminal fermentation and nutrient

absorption, a key factor in this adaptation is greatly increased ruminal surface area.

Papillary development plays an important role in stabilizing pH by preventing acid

accumulation via absorption of ruminally-produced organic acids (Dirksen et al., 1985).

The captive giraffe examined by Hofmann and Matern (1988) had approximately 11% of

the ruminal surface area of that found in wild conspecifics, suggesting they had partially

lost their ability to deal with rapid production of VFA in the rumen. Even pH reductions

that do not reach the threshold for ruminal acidosis can alter the ruminal environment,

and notable changes in fermentation characteristics occur when pH decreases to <6.2

(Russell and Wilson, 1996). If hay: concentrate intake ratios of the eleven giraffe in the

three aforementioned studies, low (relative to wild giraffe) feeding and rumination time

by giraffe in the present study, and papillae development of the two giraffe cited in

Hofmann and Matern (1988) are representative of captive giraffe in general, ruminal pH









reduction, and quite possibly some degree of ruminal acidosis, is likely occurring in a

substantial percentage of the captive giraffe population.

The lack of documentation of acidosis in captive giraffe may relate to the degree of

pH reduction and the extreme difficulty in diagnosing non-catastrophic degrees of

acidosis without rumen fluid collection (Garrett et al., 1999), which is unlikely to occur

in captive giraffe. Symptoms of subacute (pH<5.5) rather than acute (pH<5.0) acidosis

are "insidious and considerably less overt," and include decreased or variable feed intake,

decreased efficiency of milk production, poor body condition, unexplained diarrhea, and

episodic laminitis (Nocek, 1997), symptoms which have occurred in captive giraffe, but

have generally been attributed to other causes. The variation in selective consumption of

forage and concentrate noted in the present study may dictate which animals are at

greater risk of ruminal disorders. Provision of mixed diets that reduce the ability of the

animals to selectively consume concentrates may reduce this risk.

Conclusions

Captive giraffe offered EF as compared to those offered GF increased the amount

of time engaged in feeding behavior to a level closer to that observed for wild giraffe,

which also may have increased ruminal buffering via saliva production at the time of

rapidly-fermenting NFC consumption. Decreased starch and increased NDSF contents of

EF facilitated consumption of a carbohydrate profile that may bear a greater similarity to

natural giraffe feedstuffs than typical zoo concentrate diets. A possible resultant shift in

ruminal VFA production toward the high acetate: low propionate profile observed in wild

giraffe could contribute to the observed decrease in blood glucose.

The fact that few significant treatment effects were observed in this study may be a

result of low animal numbers, individual animal variation, and a relatively short time on









treatment. The experimental supplement was formulated in an attempt to improve

ruminal function and overall animal response, and while results suggest that this

formulation may have some promise, further research is needed for confirmation.


Table 3-1. Design of study.

Pen 1 Pen 2

Period Dates Giraffe Diet Giraffe Diet

1 08/20 09/09/02 2 EF 1 GF

2 09/10 09/30/02 2 GF 1 EF

3 10/01 10/21/02 2 EF 3 EF

4 11/12- 12/02/02 4 EF 3 GF

5 12/03 12/23/02 4 GF 5 GF

6 12/24 01/13/03 6 GF 5 EF

7 01/14 02/03/03 6 EF 5 GF
EF = Experimental supplement; GF = Control supplement









Table 3-2. Chemical composition of alfalfa hay and supplements (dry matter basis) fed
to captive giraffe, and difference between supplements.
-----------------------Supplement ----------
Alfalfa hay GF EF SE P-values
Dry matter (%) 90.8 88.3 85.2 0.29 0.001
Organic matter (%) 91.0 92.0 89.1 0.07 <0.001
Ash (%) 9.04 8.02 10.9 0.07 <0.001
Crude Protein (%) 19.2 17.1 17.6 0.25 0.138
NDFOM (%) 41.2 35.1 40.3 0.60 0.003
ADF (%) 33.8 23.0 26.0 0.41 0.002
Lignin (%) 7.12 3.32 5.67 0.12 <0.001
NDSF (%) 15.2 9.64 14.9 0.25 0.001
Sugar (%) 11.6 14.7 21.4 0.51 <0.001
Starch (%) 1.20 14.2 1.47 0.06 <0.001
Ca (%) 1.65 1.17 1.28 0.02 0.021
P (%) 0.22 0.79 0.65 0.02 0.007
Mg (%) 0.34 0.44 0.56 0.04 0.080
K (%) 1.63 1.21 1.91 0.03 <0.001
Na (%) 0.16 0.30 0.46 0.02 0.003
Fe (ppm) 97.8 413 674 98.9 0.003
Mn (ppm) 31.9 141 1097 75.4 <0.001
Zn (ppm) 37.6 160 282 31.0 0.050
Cu (ppm) 4.94 23.8 29.3 4.26 0.410
NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber
EF (experimental supplement) was calculated to contain 0.29% S, 0.38 ppm Co, 387 ppm Fe, 1.0 ppm I,
0.77 ppm Se, 6.29 KIU/lb Vitamin A, 1.50 KIU/lb Vitamin D,and 118.23 KIU/lb Vitamin E.
GF (control supplement) was composed of 75% Mazuri Browser Breeder and 25% Purina Omolene 200.
Mazuri Browser Breeder is reported to contain 0.21% S, 0.37 ppm Co, 370 ppm Fe, 1.0 ppm I, 0.69 ppm
Se, 16,250 IU/kg Vitamin A, 3,000 IU/kg Vitamin D3 (added), 240 IU/kg Vitamin E, 5.3 ppm Vitamin K,
12 ppm thiamin hydrochloride, 10 ppm riboflavin, 47 ppm niacin, 44 ppm pantothenic acid, 1070 ppm
choline chloride, 1.5 ppm folic acid, 9.9 ppm pyridoxine, 0.21 ppm biotin, and 46 mcg/kg Vitamin B12.
Purina Omolene 200 is reported to contain not less than 0.60 ppm Se and 3000 IU/lb Vitamin E.

Table 3-3. Effects of dietary treatment on mean daily dry matter and nutrient intake,
digestion of organic matter and crude protein (apparent) and NDFOM (true),
body weight gain and body condition score.
--------P-values--------
Intake, DM basis GF EF SE Diet Period
Total (% of BW) 1.20 1.25 0.10 0.294 0.123
Supplement (% of BW) 1.00 0.98 0.10 0.81 0.237
Hay (% of BW) 0.24 0.23 0.06 0.63 0.332
Total DM, kg/ day 7.60 7.91 0.66 0.527 0.273
Supplement, kg/ day 6.17 6.30 0.58 0.83 0.351










Table 3-3. Continued.


Intake, DM basis
Hay, kg/ day
Organic matter, kg/ day
Ash, kg/ day
Crude protein, kg/ day
NDFOM, kg/ day
Acid detergent fiber, kg/ day
Lignin, kg/ day
NDSF, kg/ day
Sugar, kg/ day
Starch, kg/ day
Calcium, kg/ day
Phosphorous, kg/ day
Magnesium, kg/ day
Potassium, kg/ day
Sodium, kg/ day
Digestibility
OM, % of DM
NDFOM, % of DM
CP, % of DM
OM, kg
NDFOM, kg
CP, kg

Body weight gain, kg
Body condition score


GF
1.47
7.08
0.52
1.32
2.80
1.83
0.33
0.85
1.12
0.93
0.10
0.05
0.03
0.10
0.02

74.1
56.0
75.5
5.09
1.52
0.97

4.84
3.85


EF
1.52
7.20
0.72
1.41
3.27
2.23
0.50
1.19
1.53
0.12
0.11
0.04
0.04
0.14
0.03

69.3
54.6
65.7
4.91
1.77
0.91

12.00
4.18


SE
0.41
0.63
0.05
0.14
0.28
0.22
0.05
0.10
0.11
0.04
0.01
0.00
0.00
0.01
0.00

1.06
1.02
2.22
0.42
0.17
0.10

3.78
0.24


--------P-values--------
Diet Period
0.80 0.348
0.725 0.239
0.157 0.351
0.136 0.095
0.198 0.287
0.039 0.078
0.064 0.275
0.074 0.251
0.115 0.272
0.052 0.675
0.048 0.042
0.185 0.329
0.176 0.201
0.013 0.057
0.112 0.219


0.206
0.513
0.151
0.143
0.077
0.253

0.328
0.247


0.413
0.331
0.593
0.063
0.124
0.194

0.454
0.515


NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber; GF = control
supplement; EF = experimental supplement

Table 3-4. Effect of diet on fecal nutrient composition (dry matter basis).
--------P-values--------
GF EF SE Diet Period
OM 62.8 64.9 6.94 0.3066 0.0181
NDFOM 42.2 44.4 6.47 0.1123 0.0011
CP (Dried feces)a 11.2 14.2 1.78 <0.0001 0.0017
CP (Wet feces)b 13.4 16.0 1.67 0.0003 0.0054
CP (Dried feces) b 11.6 14.5 1.44 0.0035 0.2830
GF = control supplement; EF = experimental supplement; NDFOM = neutral detergent fiber OM
aCombustion analysis
bKjeldahl analysis









Table 3-5. Recorded behavior of captive giraffe consuming different supplements.


---------Supplement---------


Behavior, minutes/ 48 hours
Standing
Walking
Lying
Sleeping
Social contact
Rumination
Hay consumption
Supplement consumption
Total eating
Eating + rumination
Drinking
Salt
Metal licking
Non-metal licking


GF
1567
272
1009
19
23
609
239
121
359
953
58
1
255
17


EF
1698
194
906
13
23
677
254
277
530
1224
54
0
212
22


Tongue play 161 80
Total oral stereotype 433 318
Wall foraging 37 36
Total oral behavior 1481 1632
GF = control supplement; EF = experimental supplement


SE
281
84
87
3
4
51
45
25
42
35
1
1
70
7
39
57
14
71


--------P-values--------
Diet Period


0.012
0.631


0.017
0.434


did not converge
0.374 0.475
0.994 0.034
0.307 0.322
0.762 0.431
0.063 0.255
0.100 0.312
0.124 0.187
0.416 0.589
0.416 0.589
0.390 0.357
0.716 0.493
0.119 0.302
0.223 0.549
0.983 0.816
0.363 0.455


aTotal eating + rumination + total oral stereotype + wall foraging


Table 3-6. Percentage of time female giraffe spent engaged in oral behaviors.
------------------Captivea-------- ---


Feeding
Rumination


Minimum
9.1
15.8


Maximum
38.2
32.6


Mean
17.7
22


Oral stereotypes 7.3 23 13.3
Oral behavior 43 66.1 54.1
a6 giraffe over 48 hours.
b5 giraffe during daylight hours. (Leuthold and Leuthold, 1978)


SE
2.16
1.38
1.18
2.1


Tildb


Mean
53.1
15
NA
68.1


Table 3-7. Effects of type of dietary supplement on giraffe blood parameters.
--------P-values------
Measure Units GF EF SE Diet Period
Non-esterified fatty acids mEq/L 0.47 0.38 0.06 0.282 0.283
Insulin ng/ml 1.04 0.84 0.27 0.709 0.747
Glucose mg/dl 99 82.3 8.14 0.028 0.041
Blood urea nitrogen mg/dl 20.6 16.6 0.7 0.166 0.549









Table 3-7. Continued

Measure
Creatinine
Total protein
Albumin
Total bilirubin
Alkaline phosphatase
Alanine aminotransferase
Aspartate aminotransferase
Cholesterol
Ca
P
Na
K
Cl
Mg
Globulin
Lipase
Amylase
Triglycerides
CPK
GGTP
Calculated osmolality
Lactate dehydrogenase
Hemoglobin
Hematocrit
WBC


RBC
MCV
MCH
MCHC
Neutrophils
Bands
Lymphocytes
Monocytes
Eosinophils
Basophils
Fibrinogen


Units GF


106/ul




%
%
%
%
%
%
mg/dl


11 11.2
33.1 31.9
11 11
33.7 34.2
76 76.7
0.07 0.07
14.3 11.4
2.58 4.36
6.18 5.84
1.13 1.75
244 236


0.48
0.34
0.1
0.17
4.5
0.13
2.65
0.86
1.92
0.34
11


mg/dl 1.67 1.71
g/dl 8.92 9.03
g/dl 2.49 2.49
mg/dl 0.12 0.16
U/L 158 148
U/L 11.3 7.51
U/L 55.6 45.7
mg/dl 33.2 29.2
mg/dl 8.55 8.95
mg/dl 10.9 10.2
mEq/L 147 146
mEq/L 4.58 4.86
mEq/L 107 104
mEq/L 2.17 2.07
g/dl 6.44 6.55
U/L 31.3 42.5
U/L 11.8 12.2
mg/dl 41.6 35.4
U/L 248 258
U/L 18.5 19.7
mOsm/L 296 291
U/L 401 380
g/dl 12.1 12.3
% 36 35.9
103/ul 15.2 16.5


0.831 0.553
0.263 0.284
0.765 0.277
0.262 0.454
0.843 0.46
1 0.758
0.105 0.108
0.255 0.253
0.897 0.723
did not converge
0.658 0.64


-----P-values-----


GF = control supplement; EF = experimental supplement; CPK = creatinine phosphokinase; GGTP =
gamma glutamyl transpeptidase; WBC = white blood cell; RBC = red blood cell; MCV = mean corpuscular
volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration


SE Diet
0.06 0.717
0.32 0.76
0.07 0.973
0.01 0.295
17.6 0.516
1.51 0.244
5.55 0.416
1.86 0.208
0.29 0.309
0.6 0.5
1.32 0.538
0.13 0.183
2.06 0.38
0.5 did not
0.35 0.674
5.4 0.404
0.89 0.817
5.41 0.502
69.8 0.942
3.23 0.799
3.01 0.39
23.2 0.662
0.53 0.859
1.6 0.986
1.42 0.554


Period
0.175
0.731
0.429
0.356
0.386
0.506
0.752
0.335
0.577
0.325
0.526
0.327
0.851
converge
0.833
0.425
0.551
0.874
0.737
0.511
0.523
0.544
0.694
0.7
0.58











Table 3-8. Pearson Correlation Coefficients for correlations of blood glucose and BUN
with kilograms of nutrients consumed and digested.
-----------------Kg consumed---------- -----Kg digested-----
OM NDFOMCP NDSF Sugar Starch OM* NDFOMCP*
Glucose r 0.614 0.462 0.603 0.462 0.231 0.495 0.560 0.400 0.595
P-value 0.020 0.097 0.023 0.097 0.426 0.072 0.074 0.223 0.054
BUN r 0.427 0.146 0.419 0.146 -0.253 0.816 0.521 0.202 0.583
P-value 0.128 0.619 0.136 0.619 0.382 <0.001 0.100 0.552 0.060
NDFom = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber; BUN = blood
urea nitrogen.
*Apparent digestion

Table 3-9. Pearson Correlation Coefficients for correlations of blood glucose and BUN
with various blood proteins.
Total Total Alk
protein BUN Creat CPK Albu bilirubin phos ALT AST
Glucose r -0.250 0.758 -0.303 0.525 0.231 -0.307 0.019 0.397 -0.078
P-value 0.389 0.002 0.292 0.054 0.426 0.285 0.950 0.159 0.792
BUN r -0.190 1.000 -0.374 0.515 0.212 -0.387 -0.021 0.463 0.302
P-value 0.514 0.188 0.060 0.466 0.172 0.942 0.091 0.295
BUN = blood urea nitrogen; Creat = creatinine; CPK = creatinine phosphokinase; Alk phos = alkaline
phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase











































0
GF EF EF GF EF EF GF EF GF GF EF GF GF

1 2 1 2 3 3 4 4 5 5 6 7 6

Giraffe 1 Giraffe 2 Giraffe 3 Giraffe 4 Giraffe 5 Giraff
treatment, period, giraffe

Total Oral Stereotypes Metal Licking O Tongue Pay



Figure 3-1. Number of minutes over 48 hours individual giraffe spent engaged in specific and total oral stereotype behavior.


e 6










0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
n an


II


1gf lef 2ef 2gf 2ef 3ef 3gf 4ef 4gf 5gf 5ef 5gf 6gf 6ef

Figure 3-2. Blood concentrations of non-esterified fatty acids in individual giraffe.
Giraffe are designated by number (1-6); gf = control diet; ef = experimental diet.















APPENDIX A
INDIVIDUAL ANIMAL MEASURES













Table A-1. Individual giraffe voluntary intake (DM basis).


Giraffe
Gl
Gl
G2
G2
G2
G3
G3
G4
G4
G5
00 G5
G5
G6
G6


Period
1
2
1
2
3
3
4
4
5
5
6
7
6
7


Diet
GF
EF
EF
GF
EF
EF
GF
EF
GF
GF
EF
GF
GF
EF


aForage:supplement intake ratio (DM basis)


------Percent Body Weight
Total Supplement
0.86 0.70
1.31 1.29
0.69 0.46
1.01 0.90
1.06 0.87
1.50 1.38
1.66 1.41
1.03 0.87
1.13 0.93
1.44 0.85
1.55 0.94
1.45 0.87
1.24 1.07
1.20 0.97


Alfalfa Hay
0.12
0.02
0.22
0.10
0.20
0.12
0.25
0.17
0.19
0.60
0.61
0.58
0.17
0.22


-------------Kilograms-------------
Total Supplement Alfalfa Hay
5.95 4.84 1.10
9.38 9.26 0.12
4.26 2.87 1.39
6.32 5.66 0.66
6.92 5.64 1.27
8.66 7.95 0.72
9.62 8.18 1.44
6.90 5.79 1.11
7.49 6.20 1.29
9.03 5.30 3.72
10.0 6.10 3.94
9.59 5.73 3.86
7.28 6.28 0.99
7.13 5.79 1.34


F:S Ratioa
19:81
01:99
33:67
10:90
18:82
08:92
15:85
16:84
17:83
41:59
39:61
40:60
14:86
19:81












Table A-2. Individual giraffe nutrient intake (kg) (DM basis).
Giraffe Period Diet DM Ash CP NDFOM ADF Lignin NDSF Sugar Starch Ca P Mg K Na
Gl 1 GF 5.95 0.35 1.03 2.21 1.53 0.26 0.64 0.90 0.83 0.08 0.04 0.03 0.09 0.02
G1 2 EF 9.38 1.01 1.61 4.09 2.64 0.59 1.40 1.80 0.14 0.11 0.06 0.05 0.17 0.04
G2 1 EF 4.26 0.42 0.77 1.80 1.31 0.30 0.72 0.67 0.05 0.06 0.02 0.02 0.08 0.01
G2 2 GF 6.32 0.52 1.11 2.38 1.55 0.24 0.70 0.94 0.75 0.08 0.05 0.03 0.06 0.03
G2 3 EF 6.92 0.63 1.20 2.86 1.93 0.43 1.03 1.51 0.11 0.07 0.03 0.04 0.13 0.03
G3 3 EF 8.66 0.83 1.50 3.52 2.29 0.49 1.25 1.96 0.14 0.10 0.05 0.05 0.17 0.04
G3 4 GF 9.62 0.65 1.71 3.52 2.33 0.41 1.12 1.38 1.15 0.14 0.07 0.04 0.12 0.03
G4 4 EF 6.90 0.64 1.21 2.79 1.81 0.42 1.08 1.58 0.07 0.10 0.04 0.05 0.12 0.03
G4 5 GF 7.49 0.54 1.39 2.54 1.63 0.27 0.79 1.03 1.03 0.10 0.05 0.03 0.09 0.02
G5 5 GF 9.03 0.48 1.71 3.32 2.30 0.47 1.18 1.14 0.94 0.13 0.05 0.04 0.12 0.02
G5 6 EF 10.04 0.76 1.89 4.24 3.03 0.67 1.52 1.82 0.14 0.14 0.04 0.04 0.18 0.03
G5 7 GF 9.59 0.55 1.65 3.79 2.53 0.52 1.26 1.32 0.84 0.12 0.05 0.04 0.13 0.02
G6 6 GF 7.28 0.51 1.26 2.61 1.75 0.30 0.71 0.93 1.00 0.09 0.05 0.03 0.09 0.02
G6 7 EF 7.13 0.65 1.17 2.92 2.01 0.46 1.05 1.45 0.11 0.09 0.04 0.04 0.13 0.03
NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber












giraffe fecal output, fecal nutrient concentration (% of DM), and apparent nutrient digestibility (%).


Giraffe
Gl
Gl
Gl
Gl
G2
G2
G2
G2
G2
G2
G3
G3
G4
G4
G4
G4
G5
G5
G5
G5
G5
G5
G6
G6
G6
G6


NDFOM = neutral detergent fiber organic matter


Period
1
1
2
2
1
1
2
2
3
3
3
3


5
5
5
5
6
6
7
7
6
6
7
7


Day
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2


Diet
GF
GF
EF
EF
EF
EF
GF
GF
EF
EF
EF
EF
EF
EF
GF
GF
GF
GF
EF
EF
GF
GF
GF
GF
EF
EF


Grams (DM)
2226
743
4316
3470
1813
861
2698
1783
3113
2591
3101
1245
2622
2586
2650
2630
3780
2815
4098
4078
4299
3818
2492
2981
2854
2987


----Organic matter----
Concentration Digest.
62.3 75.2
69.8 90.7
79.5 58.9
80 66.8
65 69.3
69.1 84.5
71.4 66.7
74.4 77.1
69.8 65.4
66.5 72.5
52.6 79.2
64.7 89.7
36.9 84.4
44.1 81.7
53.6 79.5
53.8 79.6
56.2 75.1
64.8 78.6
66.5 70.6
58.8 74.1
63.9 69.6
64.3 72.9
71.3 73.8
57.1 74.9
71.5 68.5
66.1 69.6


------NDFOM-------
Concentration Digest.
39.2 60.4
46.2 84.4
52.2 44.9
53.9 54.3
45.1 54.7
48.9 76.7
50.1 43.3
52.6 60.6
46.5 49.4
46.5 57.9
31.9 72.0
42.4 85.0
24.7 76.8
27.8 74.2
38.4 59.9
36.1 62.6
40.7 53.7
41.7 64.6
45.6 56.0
40.4 61.1
46.4 47.3
43.7 55.9
46.7 55.4
36.5 58.4
48.8 52.4
46.9 52.1


------Crude protein-------
Concentration Digest.
12.0 74.2
12.6 91.0
18.2 51.2
17.8 61.4
13.4 68.5
13.4 85.1
11.9 71.0
12.7 79.7
14.7 61.8
13.3 71.3
11.3 76.5
14.9 87.6
9.0 80.6
9.7 79.3
10.2 80.5
10.4 80.3
10.0 77.9
10.3 83.0
13.2 71.4
11.6 74.9
10.0 73.8
9.4 78.1
13.4 73.6
11.1 73.7
16.0 61.2
14.5 63.0


Table A-3. Individual












Table A-4. Differences in crude protein concentration of individual fecal samples analyzed in wet and dried forms.
Giraffe Diet Period Day Wet Dried Difference
Gl GF 1 1 13.89 11.95 1.93
Gl GF 1 2 16.08 13.04 3.04
Gl EF 2 1 20.51 18.96 1.55
Gl EF 2 2 20.92 18.11 2.81
G2 EF 1 1 13.84 13.47 0.37
G2 EF 1 2 15.19 13.49 1.70
G2 GF 2 1 14.13 10.24 3.89
G2 GF 2 2 14.89 13.36 1.53
G2 EF 3 1 15.61 14.86 0.75
G2 EF 3 2 13.61 13.66 -0.06
G3 EF 3 1 13.76 12.04 1.72
G3 EF 3 2 16.36 15.23 1.13
G3 GF 4 1 15.95 13.53 2.42
G3 GF 4 2 16.54 14.97 1.56
G4 EF 4 1 10.29 6.89 3.40
G4 EF 4 2 11.76 12.63 -0.88
G4 GF 5 1 12.23 10.47 1.76
G4 GF 5 2 12.32 10.83 1.49
G5 GF 5 1 12.84 10.25 2.59
G5 GF 5 2 11.89 10.86 1.03
G5 EF 6 1 15.43 13.25 2.19
G5 EF 6 2 13.60 12.18 1.42
G5 GF 7 1 11.65 10.33 1.32
G5 GF 7 2 11.70 9.79 1.91
G6 GF 6 1 15.34 13.91 1.43
G6 GF 6 2 13.92 11.83 2.10
G6 EF 7 1 17.04 16.54 0.49
G6 EF 7 2 16.34 14.86 1.48












Table A-5. Minutes over 48 hours individual giraffe spent engaged in measured behaviors.
Social Hay Supplement Total Eat +
Giraffe Period Diet Standing Walking Lying Sleep contact consumption consumption eating ruminat
Gl 1 GF 1421 151 1238 27 47 191 103 294 844
G1 2 EF 1582 191 1048 12 72 26 389 415 897
G2 1 EF 2001 78 775 6 22 233 173 406 1029
G2 2 GF 1889 59 876 14 46 149 114 263 719
G2 3 EF 1818 64 989 3 6 166 167 333 985
G3 3 EF 1954 444 472 11 15 153 144 297 860
G3 4 GF 1387 1035 409 0 7 218 137 355 837
G4 4 EF 1630 500 678 3 3 181 418 599 1232
G4 5 GF 1573 290 1015 23 23 324 223 547 1391
G5 5 GF 1906 12 859 38 32 599 155 754 1226
G5 6 EF 1802 87 990 26 13 484 265 749 1473
G5 7 GF 1706 105 1046 26 11 399 140 539 1207
G6 6 GF 1202 75 1622 23 12 221 92 313 1101
G6 7 EF 1357 112 1387 13 9 262 233 495 1434













Table A-5. Continued.


Rumination Drinking
550 70
482 57
623 5
456 4
652 5
563 145
482 184
633 26
844 86
472 27
724 38
668 33
788 17
939 11


Salt
0
0
0
0
0


Giraffe
Gl
Gl
G2
G2
G2
G3
G3
G4
G4
G5
G5
G5
G6
G6


Period
1
2
1
2
3
3
4
4
5
5
6
7
6
7


Diet
GF
EF
EF
GF
EF
EF
GF
EF
GF
GF
EF
GF
GF
EF


Metal
licking
226
189
510
640
478
86
33
220
102
240
222
389
69
209


Non-metal
licking
2
10
0
7
1
16
8
43
75
36
25
32
12
9


Tongue
play
243
49
9
14
8
149
168
165
218
9
25
18
267
131


Total oral
stereotype
471
248
519
661
487
251
209
428
395
285
272
439
348
349


Wall
foraging
97
52
0
14
2
11
7
13
32
62
49
31
48
79


Total oral
1482
1254
1553
1398
1479
1268
1237
1699
1904
1600
1832
1718
1514
1873













giraffe blood values on day 21.


Giraffe Period Diet NEFA
mEq/L
Gl 1 1 0.55
Gl 2 2 0.70
G2 1 2 0.33
G2 2 1 0.78
G2 3 2 0.17
G3 3 2 0.34
G3 4 1 0.38
G4 4 2 0.25
G4 5 1 0.41
G5 5 1 0.37
G5 6 2 0.19
G5 7 1 0.22
G6 6 1 0.57
G6 7 2 0.50


Insulin
ng/ml
0.42
0.37
1.74
1.05
0.94
0.79
2.37
0.83
1.45
0.99
0.74
0.84
0.37
0.32


GLC
mg/dl
68
92
59
118
65
105
135
87
93
89
97
103
90
64


BUN
mg/dl
17
17
15
22
16
15
24
17
22
19
19
21
20
16


Creat
mg/dl
2.3
2.3
2.3
2.1
1.9
1.6
1.4
1.6
1.4
1.3
1.3
1.5
1.3
1.3


Prot
g/dl
9.1
9.8
9.5
9.3
9.4
8.6
9.0
9.5
9.5
7.4
7.4
8.2
8.8
9.4


Albu
g/dl
2.3
2.5
2.5
2.5
2.6
2.4
2.6
2.5
2.4
2.5
2.6
3.0
2.2
2.5


TBili
mg/dl
0.1
0.1
0.2
0.1
0.2
0.2
0.1
0.2
0.2
0.2
0.1
0.1
0.1
0.1


AlkPhos ALT


U/L
223
238
130
142
117
114
160
227
149
117
109
147
124
109


U/L
7
8
5
8
5
8
17
8
8
5
6
11
19
13


NEFA = non-esterified fatty acids; GLC = glucose; BUN = blood urea nitrogen; Creat = creatinine; Prot = total protein; Albu = albumin; TBili
AlkPhos = alkaline phosphotase; ALT = alanine aminotransferase; AST = aspertate aminotransferase


AST
U/L
46
44
48
44
44
37
44
50
64
47
42
56
89
50
total bilirubin;


Table A-6. Individual












Table A-6 Continued.
GiraffePeriod Diet Ca P Na K Cl Mg Glob Lipase AmylaseTriglyc CPK GGTP Osm
mg/dl mg/dl mEq/L mEq/L mEq/L mEq/L g/dl U/L U/L mg/dl U/L U/L mOsm/L
Gl 1 GF 7.9 9.9 147 5.1 111 2.0 6.8 37 11 52 112 20 293
G1 2 EF 7.7 11.8 147 5.0 110 2.0 7.3 27 10 34 324 11 294
G2 1 EF 8.5 8.7 147 4.6 104 1.7 7.0 28 10 20 190 18 291
G2 2 GF 8.0 12.3 145 4.1 105 1.9 6.8 16 10 37 27 13 292
G2 3 EF 8.6 11.9 145 4.9 101 2.1 6.8 27 10 34 145 17 288
G3 3 EF 9.8 12.0 149 5.1 104 1.8 6.2 31 15 37 204 12 298
G3 4 GF 9.3 15.2 148 4.7 106 1.7 6.4 26 10 31 608 9 300
G4 4 EF 8.8 10.8 145 4.6 102 2.2 7.0 21 10 58 243 18 289
G4 5 GF 8.2 6.8 145 4.1 115 2.3 7.1 38 15 51 406 24 290
G5 5 GF 9.3 9.1 144 4.6 101 2.4 4.9 37 10 35 251 46 288
G5 6 EF 9.8 10.3 146 4.9 101 2.1 4.8 78 15 29 379 25 293
G5 7 GF 9.8 11.5 156 5.0 104 3.5 5.2 55 13 42 347 27 313
G6 6 GF 8.0 8.2 143 4.2 109 1.4 6.6 30 14 33 152 15 286
G6 7 EF 9.3 10.3 146 5.1 102 2.6 6.9 66 15 39 159 19 290
Glob = globulin; Triglyc = triglycerides; CPK = creatinine phosphokinase; GGTP = gamma glutamyl transpeptidase; Osm = calculated osmolality













Table A-6 Continued.
Giraffe Period Diet LactD HGB


U/L
326
350
403
362
401
315
348
325
352
442
398
462
530
448


g/dl
11.3
12.8
12.1
11.3
12.7
12.4
11.7
11.0
14.0
13.2
12.8
14.3
10.4
11.1


HCT
%
33.2
37.6
34.8
33.3
38.1
37.4
35.3
31.6
40.9
38.5
37.7
42.2
30.8
32.6


WBC RBC MCV MCH MCHC Neutph Bands Lymph Monocts Eosinphl Basophl Fibrgn


103/ul 106/ul


15.6
16.3
21.3
14.2
13.6
15.8
21.2
13.9
13.0
15.5
14.1
12.3
17.9
16.5


10.1
11.6
10.4
9.9
11.3
10.7
10.1
9.8
12.7
12.3
12.1
13.8
10.0
10.9


33.0 11.2
32.0 11.0
33.0 11.6
34.0 11.4
34.0 11.2
35.0 11.6
35.0 11.6
32.0 11.3
32.0 11.0
31.3 10.7
31.0 10.6
31.0 10.4
31.0 10.4
30.0 10.2


34.0
34.0
34.8
33.9
33.3
33.2
33.1
34.8
34.2
34.3
34.0
33.9
33.8
34.0


LactD = lactate dehydrogenase; HGB = hemoglobin; HCT = hematocrit; WBC = white blood cell; RBC
MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration


% %
9 3
15 11
3 0
16 6
15 2
6 6
10 3
14 2
11 1
24 1
19 7
25 1
11 2
12 2
red blood cell; MCV


% % mg/dl
13 1 259
7 2 218
2 251
2 281
3 3 223
10 0 200
6 2 229
9 2 243
4 1 242
12 2 242
6 3 293
5 1 248
2 0 236
2 0 215
mean corpuscular volume;





Table A-7. Body weight and body condition score of individual giraffe on days 1 and 21 of each period.
-------------Body weight (kg)------------- Body condition score
Giraffe Period Treatment Day 1 Day 28 Change Day 1 Day 28
Gl 1 GF 693 694 1 4.0 4.0
G1 2 EF 694 717 23 4.0 4.5
G2 1 EF 625 622 -4 3.5 3.5
G2 2 GF 622 630 8 3.5 3.5
G2 3 EF 630 652 23 3.5 3.5
G3 3 EF 542 559 17 3.5 4.5
G3 4 GF 577 580 2 5.0 4.5
G4 4 EF 673 668 -5 4.0 4.0
G4 5 GF 668 667 -1 4.0 4.0
G5 5 GF 611 626 15 3.5 4.0
G5 6 EF 626 651 25 4.0 4.5
G5 7 GF 651 661 10 4.5 4.5
G6 6 GF 590 589 -1 2.5 3.0
G6 7 EF 589 598 10 3.0 4.0














% of BW [] Alfalfa Hay m Supplement
1.80

1.60 -

1.40 -

1.20 -

1.00 -

0.80 -

0.60

0.40

0.20 -

0 .0 0-
0.00 ...


1 2 1 2 3 3 4 4 5 5 6 7 6

G I G2 G3 G4 G5

Figure A-1. Individual animal consumption of alfalfa hay and supplement as a percentage of body weight.

Giraffe are designated by number (G1-G6); GF = control diet; EF = experimental diet





























GF EF

1 2

Gl


GF EF GF GF EF GF GF EF

2 3 5 5 6 7 6 7

G2 G4 G5 G6
treatment, period, giraffe U OM U NDFOM E CP


Figure A-2. Individual animal digestion of neutral detergent fiber organic matter (NDFOM), and apparent digestion of OM and CP.

Giraffe are designated by number (G1-G6); GF = control diet; EF = experimental diet














APPENDIX B
CARBOHYDRATE FRACTIONING IN FEEDSTUFFS


Plant Carbohydrates


I
Cell
Contents



Organic Mono+Oligo- Starches Fructans
Acids saccharides I


I
Cell
Wall
I


Pectic Her
Substances
Galactans
P-glucans
%. I _


micelluloses Cellulose


ADF


S NDSF NDF

S Non-Starch Iolysaccharides

NDSC
Figure B-1. Carbohydrate partitioning. (Hall, 2001)

Table B-1. NDF and NFC fractions (percent of sample DM) in feedstuffs analyzed at the
University of Florida.
Feedstuff NDF Sugars Starch Soluble Fiber
Alfalfa hay 37.4 10.0 3.0 16.2
Citrus pulp 22.1 26.5 1.0 32.9
Corn meal 16.0 2.3 62.6 8.5
Cottonseed hulls <1 4.0
Soybean meal (48%) 12.8 11.2 1.5 15.9
Sugar beet pulp 44.6 12.8 0.0 30.0
Ground wheat 12.1 1.8 64.6 8.8
Data means from Hall, 2001.















APPENDIX C
INFORMATION ON CONTROL DIET COMPOSITION AND BEHAVIOR
RECORDING

Table C-1. Mean analyzed chemical composition of the batches of Purina Omelene 200
(n=5) and Mazuri Browser Breeder (n=5) fed to captive giraffe during a
giraffe feeding study.
Nutrient Purina Omelene 200 Mazuri Browser Breeder
DM, % 85.4 85.4
OM, % 93.6 93.6
Ash, % 6.37 6.37
CP, % 16.6 16.6
NDFOM, % 17.2 17.2
ADF, % 7.47 7.47
Lignin, % 1.84 1.84
NDSF, % 6.93 6.93
Sugar, % 11.0 11.0
Starch, % 38.1 38.1
Ca, % 0.68 0.68
P, % 0.56 0.56
Mg, % 0.20 0.20
K, % 1.15 1.15
Na, % 0.23 0.23
Fe, ppm 390 297
Mn, ppm 217 115
Zn, ppm 222 137
Cu, ppm 45.6 16.8
NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber
Purina Omelene 200 is a sweet feed containing: whole oats, cracked corn, ground corn, dehulled soybean
meal, corn flour, cane molasses, soybean oil, wheat middlings, dicalcium phosphate, sodium selenite,
propionic acid, dried whey, vitamin E supplement, choline chloride, citric acid, vitamin A supplement,
calcium pantothenate, ferric oxide, tocophorols, riboflavin supplement, vitamin B-12 supplement, vitamin
D3 supplement, niacin supplement, ferrous carbonate, manganous oxide, zinc oxide, copper sulfate,
magnesium oxide, ferrous oxide, calcium iodate, cobalt carbonate, DL-methionine, L-lysine.
Mazuri Browser Breeder is a pelleted concentrate containing: ground soybean hulls, wheat middlings,
ground aspen, dehulled soybean meal, dried beet pulp, cane molasses, dehydrated alfalfa meal, sucrose,
brewers dried yeast, soybean oil, dicalcium phosphate, salt, calcium carbonate, magnesium oxide,
menadione dimethylpyrimidinol bisulfite (source of vitamin K), pyridoxine hydrochloride, dl-alpha
tocopheryl acetate (source of vitamin E), cholecalciferol (source of vitamin D3), vitamin A acetate, calcium
pantothenate, ethoxyquin (a preservative), thiamin mononitrate, cyanocobalamin (source of vitamin B12),
riboflavin, biotin, nicotinic acid, manganous oxide, zinc oxide, ferrous carbonate, copper sulfate, zinc
sulfate, calcium iodate, cobalt carbonate, sodium selenite.









Table C-2. Behavioral category definitions used by observers when recording giraffe
behaviors.
Cud Chewing cud (ruminating)
Supp Eating supplement from feed pan
Hay Eating hay from hay rack
Drink Drinking
Salt Licking salt block
Dirt Eating dirt
Metal Licking metal
Lick Licking or mounting non-metal inanimate objects
Wall Foraging in or mouthing area behind back wall
Tongue Tongue playing not associated with licking, eating, drinking, or ruminating
Social Making physical contact with other giraffe
Stand Standing
Walk Walking
Lay Lying down with head up
Sleep Sleeping lying down with head folded over onto flank





















-- lI'. 1.1 ll ).I ti i l --I ll II.... .. l Ii..l *i'. ,11 .l 1













I IIi









II I
II 11









II I



II I
II Il'














Ii I I
I I 4i
1 1 1 -1


















































II 44
















iii 4 _
ii I


I.ni


. ale o daa d d ia ai.


Figure C-I. Example of data sheet used to record giraffe behavior.


t mI ai l 1 I ".1 i. "


II, I..I. 1 4 10'


1 1 i 1 I oI I


i i i t ;- H ... i .. I .|.-. .,.l















LIST OF REFERENCES


Allen MS. 1996. Physical constraints on voluntary intake of forages by ruminants. J
Anim Sci 74:3063-3075.

Allen MS. 1997. Relationship between fermentation acid production in the rumen and the
requirement for physically effective fiber. J Dairy Sci 80:1447-1462.

Ariza P, Bach A, Stem MD, Hall MB. 2001. Effects of carbohydrates from citrus pulp
and hominy feed on microbial fermentation in continuous culture. J Anim Sci
79:2713-2718.

Association of Official Analytical Chemists. 1990. Official Methods of Analysis, 15th
edition. Arlington, VA.

Baer DJ, Oftedal OT, Fahey Jr. GC. 1985. Feed selection and digestibility by captive
giraffe. Zoo Biol 4:57-64.

Bailey CB. 1961. Saliva secretion and its relation to feeding in cattle. III. The rate of
secretion of mixed saliva in the cow during eating, with an estimate of the
magnitude of the total daily secretion of mixed saliva. Br J Nutr 15:443-451.

Ball RL, Kearney C, Burton M, Dumoneux G, Olsen JH. 2002. Morbidity and mortality
related to hypoglycemia and chronic energy malnutrition in captive giraffe. In: Proc
Am Assoc Zoo Vets, p 180-185.

Barnes R, Greene K, Lamm M. 2002. Management and husbandry of duikers at the Los
Angeles Zoo. Zoo Biol 21:107-121.

Bashaw MJ, Tarou LR, Maki TS, Maple TL. 2001. A survey assessment of variable
related to stereotypy in captive giraffe. Appl Anim Behaviour Sci 73:235-247.

Beauchemin KA, Yang WZ, Rode LM. 2003. Effects of particle size of alfalfa-based
dairy cow diets on chewing activity, ruminal fermentation, and milk production. J
Dairy Sci 86:630-643.

Beharka AA, Nagaraja TG, Morrill JL, Kennedy GA, Klemm RD. 1998. Effects of form
of diet on anatomical, microbial, and fermentative development of the rumen of
neonatal calves. J Dairy Sci 81:1946-1955.









Ben-Ghedalia D, YosefE, Miron J, Est Y. 1989. The effects of starch- and pectin-rich
diets on quantitative aspects of digestion in sheep. Anim Feed Sci Technol 24:289-
298.

Bernard L, Chaise JP, Baumont R, Poncet C. 2000. The effect of physical form of
orchardgrass hay on the passage of particulate matter through the rumen of sheep. J
Anim Sci 78:1338-1354.

Bhattacharya AN, Lubbadah WF. 1971. Feeding high levels of beet pulp in high
concentrate dairy rations. J Dairy Sci 54:95-99.

Burton MS, Dierenfeld ES. 1990. Comparisons of vitamin E injectables in giraffes
(Giraffa Camelopardalis reticulata). In: Proc Am Assoc Zoo Vets, p 198-201.

Caister LE, Shields WM, Gosser A. 2003. Female tannin avoidance: a possible
explanation for habitat and dietary segregatoin of giraffes (Giraffa camelopardalis
peralta) in Niger. Afr J Ecol 41:201-210.

Callison SL, Firkins JL, Eastridge ML, Hull BL. 2001. Site of nutrient digestion by dairy
cows fed corn of different particle sizes or steam-rolled. J Dairy Sci 84:1458-1467.

Campbell CP, Marshall SA, Mandell IB, Wilton JW. 1992. Effects of source of dietary
neutral detergent fiber on chewing behavior in beef cattle fed pelleted concentrates
with or without supplemental roughage. J Anim Sci 70:894-903.

Campling RC, Balch CC. 1961. Factors affecting the voluntary intake of food by cows. 1.
Preliminary observations on the effect, on the voluntary intake of hay, of changes
in the amount of the reticulo-ruminal contents. Br J Nutr 15:523-530.

Cassida KA, Stokes MR. 1986. Eating and resting salivation in early lactation dairy
cows. J Dairy Sci 69:1282-1292.

Chaffe PS. 1968. Report on death of a giraffe. J Small Anim Pract 9:136-137.

Chai K, Kennedy PM, Milligan LP. 1984. Reduction in particle size during rumination in
cattle. Can J Anim Sci 64(Suppl.):339-340.

Chester-Jones H, Metwally HM, Linn JG, Ziegler DM, Stern MD. 1991. Effects of
dietary protein-energy interrelationships on Holstein steer performance and ruminal
bacterial fermentation in continuous culture. J Anim Sci 69:4956-4966.

Ciofolo I, LePendu Y. 2002. The feeding behaviour of giraffe in Niger. Mammalia
66:183-194.

Clark PW, Armentano LE. 2002. Influence of particle size on the effectiveness of the
fiber in alfalfa silage. J Dairy Sci 85:3000-3007.









Clauss M. 1998. Feeding Giraffe (Giraffe camelopardis) [Masters thesis]. London, UK:
University of London. 114 p.

Clauss M, Kienzle E, Wiesner H. 2002a. Importance of the wasting syndrome complex in
captive moose (Alces alces). Zoo Biol 21:499-506.

Clauss M, Lechner-Doll M. 2001. Differences in selective reticulo-ruminal particle
retention as a key factor in ruminant diversification. Oecologia 129:321-327.

Clauss M, Lechner-Doll M, Flach EJ, Tack C, Hatt JM. 2001. Comparative use of four
different marker systems for the estimation of digestibility and low food intake in a
group of captive giraffes (Giraffa camelopardalis). Zoo Biol 20:315-329.

Clauss M, Lechner-Doll M, Flach EJ, Wisser J, Hatt J-M. 2002b. Digestive tract
pathology of captive giraffe (Giraffa camelopardalis) an unifying hypothesis.
European Association of Zoo and Wildlife Veterinarians Fourth Scientific Meeting.
Heidelberg, Germany. p 99-107.

Clemens ET, Maloiy GMO, Sutton JD. 1983. Molar proportions of volatile fatty acids in
the gastrointestinal tract of East African wild ruminants. Comp Biochem Physiol
76A:217-224.

Dagg AI, Foster JB. 1976. The Giraffe. Its Biology, Behavior, and Ecology. Malabar, FL:
Krieger Publishing Company.

Dehority BA, Odenyo AA. 2003. Influence of diet on the rumen protozoal fauna of
indigenous African wild ruminants. J Eukaryot Microbiol 50:220-223.

Dierenfeld ES, Mueller PJ, Hall MB. 2002. Duikers: Native food composition,
micronutrient assessment, and implications for improving captive diets. Zoo Biol
21:185-196.

Dierenfeld ES, Wildman REC, Romo S. 2000. Feed intake, diet utilization, and
composition of browses consumed by the Sumatran rhino (Dicerorhinus
sumatrensis) in a North American zoo. Zoo Biol 19:169-180.

Dirksen GU, Liebich HG, Mayer E. 1985. Adaptive changes of the ruminal mucosa and
their functional and clinical significance. The Bovine Practitioner 20:116-120.

Ditchkoff SS. 2000. A decade since "diversification of ruminants": has our knowledge
improved? Oecologia 125:82-84.

Enqvist KE. 2003. Dental disease and serious atrophy of fat syndrome in captive giraffes
(Giraffa camelopardalis). Minneapolis, MN: Annual conference of the American
Association of Zoo Veterinarians.

Field CR, Ross IC. 1976. The savanna ecology of Kidepo Valley National Forest. E Afr
Wildl J 14:1-15.









Firkins JL. 1997. Effects of feeding nonforage fiber sources on site of fiber digestion. J
Dairy Sci 80:1426-1437.

Flach EJ, Thornton SM, Kirkwood JK, Sainesbury AW. 1997. Chronic loss of condition
with persistent neutrophilia in a reticulated giraffe (Giraffa camelopardalis). In:
Proc British Vet Zoo Soc, p 33-37.

Foose TJ. 1982. Trophic strategies of ruminant versus nonruminant ungulates [PhD
Dissertation]. Chicago, IL: The University of Chicago. 334 p.

Foster JB. 1966. The giraffe of Nairobi National Park: home range, sex ratios, the herd
and food. E Afr Wildl J 4:139-148.

Fowler ME. 1978. Peracute mortality in captive giraffe. J Am Vet Med Assoc 173:1088-
1093.

Fox H. 1938. The giraffe. Some notes upon the natural characters of this animals, its care
and its misfortune. In: Zso, editor. Report of the Penrose Research Laboratory,
formerly the Laboratory and Museum of Comparative Pathology. Philadelphia, PA:
Zoological Society of Philadelphia. p 35-67.

Friggens N, Emmans GC, Robertson S, Chamberlain DG, Whittemore CT, Oldham JD.
1995. The lactational responses of dairy cows to amount of feed and to the source
of carbohydrate energy. J Dairy Sci 78:1734-1744.

Friggens NC, Oldham JD, Dewhurst RJ, Horgan G. 1998. Proportions of volatile fatty
acids in relation to the chemical composition of feeds based on grass silage. J Dairy
Sci 81:1331-1344.

Furstenburg D, Van Hoven W. 1994. Condensed tannin as an anti-defoliate agent against
browsing by giraffe (Giraffe camelopardalis) in the Kruger National Park. Comp
Biochem Physiol 107A(2):425-431.

Garrett EF, Pereira MN, Nordlund KV, Armentano LE, Goodger WJ, Oetzel GR. 1999.
Diagnostic methods for the detection of subacute ruminal acidosis in dairy cows. J
Dairy Sci 82:1170-1178.

Glock RD, DeGroot BD. 1998. Sudden death of feedlot cattle. J Anim Sci 76:315-319.

Goering HK, Van Soest PJ. 1970. Forage fiber analysis (apparatus, reagents, procedures
and some applications). In: Agric. Handb. No. 379. Washington, DC: ARS-USDA.

Gordon IJ, Illius AW. 1996. The nutritional ecology of African ruminants: a
reinterpretation. J Anim Ecol 65:18-28.

Grant RJ. 1997. Interactions among forages and nonforage fiber sources. J Dairy Sci
80:1438-1446.









Hall MB. 1999. Management strategies against ruminal acidosis. In: Tenth Annual
Florida Ruminant Nutrition Symposium. Gainesville, FL: University of Florida. p
104-113.

Hall MB. 2001. Neutral Detergent-Soluble Carbohydrates, Nutritional Relevance and
Analysis, A Laboratory Manual. Gainesvelle: University of Florida Report nr 339.

Hall MB, Herejk C. 2001. Differences in yields of microbial crude protein from in vitro
fermentation of carbohydrates. J Dairy Sci 84:2486-2493.

Hall MB, Hoover WH, Jennings JP, Miller Webster TK. 1999. A method for partitioning
neutral detergent-soluble carbohydrates. J Sci Food Agric 79:2079-2086.

Hall-Martin AJ, Skinner JD, Van Dyk JM. 1975. Reproduction in the giraffe in relation to
some environmental factors. E Afr Wildl J 13:237-248.

Hatt JM, Lechner-Doll M, Mayes B. 1998. The use of dosed and herbage n-alkanes as
markers for the determination of digestive strategies of captive giraffes (Giraffa
camelopardalis). Zoo Biol 17:295-309.

Heldt JS, Cochran RC, Stokka GL, Farmer CG, Mathis CP, Titgemeyer EC, Nagaraja
TG. 1999. Effects of different supplemental sugars and starch fed in combination
with degradable intake protein on low-quality forage use by beef steers. J Anim Sci
77:2793-2802.

Hofmann RR. 1973. The Ruminant Stomach--Stomach Structure and Feeding Habits of
East African Game Ruminants. Odhiambo TR, editor. Nairobi, Kenya: East African
Literature Bureau.

Hofmann RR. 1984. Comparative anatomical studies imply adaptive variations of
ruminants digestive physiology. J Anim Sci 64:203-205.

Hofmann RR. 1988. Morphophysiological evolutionary adaptations of the ruminant
digestive system. In: Dobson A, Dobson MJ, editors. Aspects of Digestive
Physiology in Ruminants: Proceedings of a Satellite Symposium of the 30th
International Congress of the International Union of Physiological Sciences. p 1-
20.

Hofmann RR, Kock RA, Ludwig J, Axmacher H. 1988. Seasonal changes in rumen
papillary development and body condition in free ranging Chinese water deer
(Hydropotes inermis). J Zool 216:103-117.

Hofmann RR, Matern B. 1988. Changes in gastrointestinal morphology related to
nutrition in giraffes. Cons Sci and Zoos 22:168-176.

Horwitz W. 1982. Evaluation of analytical methods used for regulation of food and
drugs. Anal Chem 54:67A-76A.




Full Text

PAGE 1

EFFECTS OF DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE ON CAPTIVE GIRAFFE By CELESTE C. KEARNEY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Celeste C. Kearney

PAGE 3

This work is dedicated, with gratitude to the Maker of all living things. This thesis is dedicated to my family, near and far. It is especially dedicated to My father, who shared his deep, abiding se nse of respect and wonder for all Creation My mother, who encouraged me to pursue my goals and fight for what I believe My siblings, natural and “adopted,” who offered love and support My nieces and nephews, a constant source of joy and pride And most of all my loving husband, Who made me complete.

PAGE 4

iv ACKNOWLEDGMENTS This research was a team effort. Th e development and execution of this study would not have been possible without th e generous contributions of numerous individuals. My sincerest th anks go out to my committee chai r, Dr. Mary Beth Hall, and committee members, Drs. Ellen Dierenfeld, Lee McDowell, and Charles Staples, for assistance and instruct ion throughout my graduate program, and for taking a chance on an unusual project; Busch Entertainment Corporation, for permitting and sponsoring the study; Dr. Ray Ball, Giraffe SSP Veterinary Advisor, for initial instruction in giraffe health and nutrition, and continued selfless as sistance through all phase s of this research; Dr. Ramon Littell, University of Florida St atistics Department, for assistance with statistical design of the study; Dr. Judy St Ledger, BEC Director of Veterinary Pathology, for encouragement and considerable intellectual contributions; the preceding researchers who have taken an interest in th e nutrition of captive concentrate selectors, particularly Dr. Marcus Clauss, for hi s literature donations and thoughtful e-mail discussions; Heidi Bissell, Amanda Dinges and Marti Roberson, for volunteer giraffe observations; Kellie and the Robersons, for food, housing and helpful distractions during the study; Alexandra Amorocho, Heidi Bissel l, Faith Cullens, Colleen Larson, and especially Lucia Holtshausen and Ashley H ughes, students of the UF Dairy Nutrition Lab, for assistance with feed mixing a nd laboratory analyses; John Funk, Jocelyn Jennings, Jan Kivipelto, Sergei Sennikov a nd Nancy Wilkinson, UF Animal Sciences technicians, for training and assistance with sample preparation a nd laboratory analyses;

PAGE 5

v and the staff of the UF Dairy Research Unit (Hague, FL), for assistance in mixing the experimental ration. Last but certainly far from least, I would like to acknowledge and thank the following staff at Busch Gardens, Tampa: Chris Bliss, for passing along the knowledge and experience of generations; Ch ris Allen, Joaquin Alonso, Kellie Anderson, Kathy Driggers, James Hammerton, Brian Hart Waylon Kerr, Chris Merrifield, Charles Moss, Jennifer Phelps, April Richardson, Pandora Sokol, Bobby Toomy and Jerry Washburn, hoofstock keepers, for long days a nd late nights of hard work during the collection phase of this labor-intensive study; Richard Baker, Alan Cross, Cindy Davis, Andrea Demuth, Joe Devlin, Dr. John Ol sen and Glenn Young, zoo and hoofstock management, for logistical coordination and providing the essential animal, facility, and labor resources; Dr. Mike Burton, Dr. Ge nevieve Demonceaux, Heather Henry, Ian Hutchinson, Cliff Martel and Mary Port, veterinary and hospital staff, for assistance with sample collection and analysis; and the entire Busch Gardens Zoo staff, for their ongoing commitment to animal care, and for their thoughts, assistance and encouragement during this project.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.............................................................................................................x ABSTRACT....................................................................................................................... xi CHAPTER 1 INTRODUCTION........................................................................................................1 2 REVIEW OF THE LITERATURE ON DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE IN RUMINANT DIETS............................................8 Physical Form...............................................................................................................8 Effects of Particle Size on Masti cation, Saliva Flow, and Ruminal pH................9 Effects of Particle Size on Digesta Passage, Intake, and Fermentation..............13 Potential Implications of peNDF for Captive Giraffe.........................................18 Dietary Carbohydrate Profile......................................................................................20 Carbohydrate Fractions.......................................................................................20 Proximate analysis system...........................................................................21 Detergent system..........................................................................................21 Carbohydrates in Natural CS Diets.....................................................................21 Effects of NFC Source on Fermenta tion Characteristics and Animal Performance.....................................................................................................24 Interaction Between Dietary Components...........................................................28 Potential Implications of Dietar y NFC Profile for Captive Giraffe....................29 3 EFFECTS OF ALTERING THE P HYSICAL FORM AND CARBOHYDRATE PROFILE OF THE DIET ON CAPTIVE GIRAFFE.................................................36 Introduction.................................................................................................................36 Materials and Methods...............................................................................................38 Design..................................................................................................................38 Giraffe..................................................................................................................39

PAGE 7

vii Facilities..............................................................................................................39 Diets.....................................................................................................................40 Sample Collection and Analyses.........................................................................41 Feedstuffs and intake....................................................................................41 Fecal collection and analysis........................................................................43 Behavior.......................................................................................................45 Body weight and blood samples...................................................................46 Statistical analysis........................................................................................47 Results and Discussion...............................................................................................47 Intake...................................................................................................................48 Digestibility.........................................................................................................49 Behavior..............................................................................................................51 Blood Measures...................................................................................................53 Ancillary Study Observations / Individual Animal Effects.................................54 Diet Selection......................................................................................................56 Conclusions.........................................................................................................58 APPENDIX A INDIVIDUAL ANIMAL MEASURES .................................................................... 68 B CARBOHYDRATE FRACTIONING IN FEEDSTUFFS.........................................80 C INFORMATION ON CONTROL DIET COMPOSITION AND BEHAVIOR RECORDING.............................................................................................................81 LIST OF REFERENCES...................................................................................................84 BIOGRAPHICAL SKETCH.............................................................................................94

PAGE 8

viii LIST OF TABLES Table page 2-1 Effects of particle size of alfalfa-ba sed dairy cow diets on chewing activity and ruminal pH................................................................................................................32 2-2 Effects of concentrate le vel and feeding management on ruminal pH of lactating dairy cows................................................................................................................32 2-3 Effects of forage particle size and grain fermentability on chewing activity, ruminal pH, and ruminal VFA profile in midlactation cows...................................33 2-4 Nutrient intake and digestion coeffi cients from giraffe fed all-hay diets.................33 2-5 Chemical composition (DM basis) of five browse plants grown at Busch Gardens in Tampa, Florida.......................................................................................34 2-6 Influence of supplemental carbohydrate source fed in combination with 0.122% BW/ day of degradable intake protein on ruminal fermentation characteristics...........................................................................................................35 3-1 Design of study.........................................................................................................59 3-2 Chemical composition of alfalfa hay a nd supplements (dry matter basis) fed to captive giraffe, and difference between supplements..........................................60 3-3 Effects of dietary treatment on mean daily dry matter and nutrient intake, digestion of organic matter and crude protein (apparent) and NDFOM (true), body weight gain and body condition score.............................................................60 3-4 Effect of diet on fecal nutrie nt composition (dry matter basis)................................61 3-5 Recorded behavior of captive gira ffe consuming different supplements.................62 3-6 Percentage of time female giraffe spent engaged in oral behaviors.........................62 3-7 Effects of type of dietary s upplement on giraffe blood parameters.........................62 3-8 Pearson Correlation Coefficients fo r correlations of blood glucose and BUN with kilograms of nutrients consumed and digested................................................64

PAGE 9

ix 3-9 Pearson Correlation Coefficients fo r correlations of blood glucose and BUN with various blood proteins......................................................................................64 A-1 Individual giraffe volunt ary intake (DM basis)........................................................68 A-2 Individual giraffe nutrien t intake (kg) (DM basis)...................................................69 A-3 Individual giraffe fecal output, fecal nutrient concentration (% of DM), and apparent nutrient digestibility (%)............................................................................70 A-4 Differences in crude protein concentra tion of individual feca l samples analyzed in wet and dried forms..............................................................................................71 A-5 Minutes over 48 hours individual giraffe spent engaged in measured behaviors....72 A-6 Individual giraffe blood values on day 21................................................................74 A-7 Body weight and body condition score of individual giraffe on days 1 and 21 of each period...............................................................................................................77 B-1 NDF and NFC fractions (percent of samp le DM) in feedstuffs analyzed at the University of Florida................................................................................................80 C-1 Mean analyzed chemical composition of the batches of Purina Omelene 200 (n=5) and Mazuri Browser Breeder (n=5) fed to captive giraffe during a giraffe feeding study............................................................................................................81 C-2 Behavioral category definitions used by observers when recording giraffe behaviors..................................................................................................................82

PAGE 10

x LIST OF FIGURES Figure page 3-1 Number of minutes over 48 hours individua l giraffe spent engaged in specific and total oral stereotype behavior............................................................................65 3-2 Blood concentrations of non-esterified fatty acids in individual giraffe..................66 A-1 Individual animal consumption of alfa lfa hay and supplement as a percentage of body weight..............................................................................................................78 A-2 Individual animal digestion of neutra l detergent fiber organic matter (NDFOM), and apparent digestion of OM and CP.....................................................................79 B-1 Carbohydrate partiti oning. (Hall, 2001)..................................................................80 C-1 Example of data sheet used to record giraffe behavior............................................83

PAGE 11

xi Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science EFFECTS OF DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE ON CAPTIVE GIRAFFE By Celeste C. Kearney May, 2005 Chair: Mary Beth Hall Major Department: Animal Sciences The effects of altering physical form and carbohydrate profile of giraffe diets were evaluated using six non-lactati ng adult female giraffe in a modified reversal study. Dietary treatments consisted of a supple ment ration composed of commonly fed commercial concentrates (GF) and an experi mental supplement (EF) containing greater concentrations of sugars and so luble fiber and lesser concentra tions of starch than GF, as well as small, heavily lignified particles used to modify dietary fiber size and texture. Each study animal was housed individually and fed EF or GF ad libitum for 21 days, and then received the other feed supplement in the subsequent 21 day period. Alfalfa hay, salt and water were offered ad libitum in all periods. In each period, blood samples were collected before feeding on day 21, feed refu sals and fecal samples were collected on days 15 through 21, and behavior was r ecorded for 48 hr via observation and instantaneous sampling on days 13 through 15. Feed intake, blood measures, and minutes spent exhibiting various behaviors were evaluated.

PAGE 12

xii Data were analyzed with a statistical m odel that included animal, period, and diet. Data presented are least squares mean s. Significance was declared at P <0.10 and tendency at 0.10< P <0.15. Blood glucose (mg/dl) was lower in animals consuming EF than GF. Average daily DM intake varied greatly among animals for both alfalfa hay (0.12 to 3.94 kg/day) and supplement (2.87 to 9.26 kg/day), but did not differ between diets. Starch intake by giraffe decrease d from 0.92 kg/day on GF to 0.12 kg/ day EF, sugar intake tended ( P =0.115) to increase from 1.12 kg/day on GF to 1.53 kg/day on EF, and neutral detergent-soluble fiber (NDSF) intake increased from 0.85 kg/day on GF to 1.19 kg/day on EF. Time engaged in supplem ent consumption was greater on EF than GF and total feed consumption + rumination time tended to be greater on EF than GF, which may have increased saliva flow and buffering of the rumen. Despite few animals and high variability in their f eed selection and intake, the da ta suggest that EF facilitated small but measurable changes in animal res ponse. Further investigation with a larger population of animals is needed.

PAGE 13

1 CHAPTER 1 INTRODUCTION Numerous health problems that are suspected to be of nutritional origin have been documented in captive giraffe. Pathologies that may relate to vitamin and mineral intake or metabolism include white muscle diseas e (Strafuss, 1973; Bu rton, 1990), urolithiosis (Wolfe et al., 2000), and dental disease (E nqvist, 2003). Pancreatic pathologies ( Fox, 1938; Fowler, 1978; Lechowski et al., 1991; Ball et al., 2002), decreased ruminal absorptive surface area (Hofmann and Mater n, 1988), ruminal acidosis (Clauss, 1998; Clauss et al., 2002b), fermentative gastritis or rumenitis (Fox, 1938; Ball et al., 2002) and gastrointestinal ulceration (F ox, 1938; Fowler, 1978) also have been documented. Firstyear calf mortality may be as high as 45% (Lackey and LaRue, 1997). Failure of passive transfer (Miller et al., 1996), calf mortality du e to poor milk intake (Flach et al., 1997), and anecdotal reports of calf mortality or hand-rearing due to “maternal failure” may relate to low colostrum and milk production due to poor nutritional stat us of giraffe dams. Wasting and sudden death (Fox, 1938; Chaffe, 1968; Fowler, 1978; St randberg et al., 1984; Junge and Bradley, 1993; Flach et al ., 1997; Ball et al., 2002) are frequently reported in the literature and anecdotally. At this time, the true proportion of captive giraffe mortality caused by nutri tional pathologies is unknown. The term “Peracute Mortality Syndrome” (PMS) was used to describe giraffe wasting/ sudden death by Dr. Murray Fowl er in 1978, following four cases of sudden death at the Sacramento Zoo. Concurrent disease (tuberculosis, treated with isoniazid powder) occurred in all four giraffe; weight lo ss despite reportedly adequate intake and

PAGE 14

2 recent (less than one month prior to death) parturition occurred in two animals. Necropsies were performed on th ree of the four animals. Absence of perirenal fat and generalized serous atrophy of adipose tissu e (3 of 3 giraffe) and marked pancreatic atrophy (2 of 3) were notable findings. In a subsequent survey, 14 of 42 responding institutions reported unexplaine d deaths of giraffe or subm itted necropsy reports listing findings consistent with the Sacramento Zoo cases. Peracute death, emaciation, concurrent disease or stress ep isode, serous atrophy of ad ipose tissue, pulmonary edema and trachial froth, petechial hemorrhage of serosal surfaces, and gastrointestinal ulceration were common findings. Fowler hypothesized that both chronic pr edisposing factors and acute trigger episodes contributed to the occu rrence of PMS. Based on the information available at the time, chronic protein or energy deficiencies were listed as the most likely predisposing factors. It was recommended that giraffe be offered low fiber diet s containing 15 to 18% CP for adult non-lactating animals, and 18 to 20% CP for calves and lactating cows. However, a 1993 follow-up study by Junge and Bradley reported nine additional cases of PMS in giraffe offered diets meeti ng these protein recommendations. Chronic energy deficiency has agai n come under scrutiny as a possible predisposing factor for PMS (Ball et al., 2002) Consistent findings of depletion and serous atrophy of adipose tissue stores (F ox, 1938; Chaffe, 1968; Strafuss and Kennedy, 1973; Fowler, 1978; Strandberg et al., 1984; Junge and Brad ley, 1993; Ball et al., 2002) are indicative of negative energy balance. When energy expenditure exceeds available dietary energy, a catabolic state occurs, a nd body reserves are mobilized. Once body fat stores are excessively depleted, rapid catabolis m of even the smallest amounts of adipose

PAGE 15

3 tissue takes place in an attempt to meet energy demands. The result is serous atrophy of adipose tissue (Smith et al., 1972). Acute energy deficiency has been proposed as an immediate cause of death in giraffe PMS. Ball et al. (2002) reported on the rapid wasting and mortality of two female giraffe (A and B) during the third trimester of first pregnancy. Anteand postmortem findings were consistent with PMS. Histopathologic findings included serous atrophy of mesenteric and epicardial fat, lymphohistio cytic rumenitis and pulmonary congestion in both cases. Pancreatic atrophy and generalized muscle atrophy were also noted in giraffe B. Blood glucose concentrations were 20 mg/d l at the time of death in giraffe A, and 3 mg/dl at 5 hours post-mortem in giraffe B. Hypoglycemia, caused by depletion of body reserves followed by an acute stressor (parturition), was pr oposed as the immediate cause of death. Since this publication, blood sample s collected within 20 minutes of death have revealed glucose levels of 6 and 12 mg/dl in two giraffe succumbing to PMS, and 297 mg/dl at less than 1 hour post-mortem in a giraffe expiring from an observed cervical injury (R. Ball, personal communications). It should be noted that endemic pathologi es of unknown or suspected nutritional origin are not isolated to captive giraffe. Wasting and mortality from unknown causes, and from digestive pathologies, have been re ported in other captive concentrate selectors (CS) (Paglia and Miller, 1992; S hochat et al., 1997; Dierenfeld et al., 2000; Clauss et al., 2002a; Willette et al., 2002). High (30 to 40% in some collections) neonatal mortality, wasting syndrome, ruminal hypomotility syndrome, bloat, and rumenitis have been described as prevalent but underreported in captive duikers (Willette et al., 2002), and hand-rearing, diet modification, and brow se supplementation were among factors

PAGE 16

4 associated with improved health and increased (up to 2x) lifespan in one institution (Barnes et al., 2002). Moose, the second larges t ruminant concentrate selector (CS), are rarely exhibited in zoos because of premat ure mortality (Shochat et al., 1997). Moose “wasting syndrome complex,” a syndrome of suspected nutritional origin, was the diagnosed cause of death in 47% of 131 adult mortalities (Clauss et al., 2002a). Sudden death, frequently attributed to digestive disord ers such as ruminal acidosis and bloat, is a common occurrence in feedlot cattle (Glock and DeGroot, 1998). While the importance of nutrition in main taining health, welf are, and reproductive status of captive wildlife is receiving increasing recognition, the research that can be performed using captive exotic animals is lim ited by a number of factors. The number of animals of a given species housed in a single institution is often small, and collections typically consist of animals in different physiological states (growth, pregnancy, lactation), making it difficult to obtain a sufficient number of similar research animals. Since many of the species housed in zoologica l institutions are rare or endangered, their conservation value prohibits placing them in potentially harmful situations. Thus, the herd instincts and fearful or aggressive temperament of many captive ungulates limits collection of data that require s individual housing or animal-h uman contact. As a result, statistically viable data on the effects of diet modification of i ndividually or group-fed animals is difficult to obtain. At the present time, in-depth nutrition al knowledge of CS ruminants remains scarce. The true nutritional requirements of giraffe and other captive CS and dietary factors contributing to suspected nutrition -related pathologies remain unquantified. Because of the dearth of data on the ingr edient and chemical composition of diets

PAGE 17

5 consumed in the wild and on nutrient requi rements of exotic ruminants, domestic ruminants have been used as models for ration formulation. However, numerous differences between domestic and wild ru minants must be considered. While the objective of ration formulation for most dom estic ruminants is to optimize relatively short-term production, the objective of ration formulation for captive wildlife is to maximize longevity and long-term health and reproduction, which may extend into decades of life. Furthermore, differen ces in digestive morphology and physiology may create discrepancies in how dietary items ar e utilized. When ruminants are classified according to natural diet and digestive anat omy, domestic cattle and sheep are grazers (GR), adapted to consumption of a predominan tly grass diet. The largest living ruminant, the giraffe, is a CS, consuming primarily foliage in its natural environment. Over 40% of the approximately 150 known extant ruminant sp ecies are classified as CS (Wood et al., 2000) and consume little or no grass, subsisting instead on fruit and foliage from trees, shrubs, and herbs. Differences between wild CS and domestic GR include not only dietary constituents, but also rumen micr obial population (Dehority and Odenyo, 2003) and gastrointestinal an atomy (Hofmann, 1973, 1984). Digestive morphophysiological differences between ruminant CS and GR have been widely documented and discussed (H ofmann, 1973; Kay et al., 1980; Gordon and Illius, 1996; Shipley, 1999; Ditchkoff, 2000) and a review is presented in the proceedings of the 30th International Congress of the International Un ion of Physiological Sciences (Hofmann, 1988). A br ief summary of differences pertinent to the research presented in this thesis follows. Generally, CS have proportionally larger salivary glands than GR. Parotid glands, for example, ra nge from 0.18 to 0.25% of body weight (BW) in

PAGE 18

6 CS, but only 0.05 to 0.07% of BW in GR. The su spected increase in saliva flow in CS may or may not facilitate increased ruminal buffering, since the apparently welldeveloped ventricular groove may allow a proportion of swallowed saliva to bypass the rumen. Such a bypass mechanism would also allow increased amounts of dietary cell solubles to escape ruminal fermentation. Th e abundance of intestinal Na+/glucose cotransporter in the brush border membrane of wild moose and roe deer is suggestive of some mechanism of ruminal escape for diet ary sugars (Rowell-Schafer et al., 1999; Wood et al., 2000). Decreased ruminal compar tmentalization and incr eased diameter of the reticulo-omasal orifice in CS (Hof mann, 1973) may contribute to the observed increased rate of digesta passage (Clauss, 1998) while the larger capacity of the lower GI tract suggests greater reliance on hind-gut di gestion. Decreased rumen capacity, greatly increased ruminal surface area due to dense, even papillation, and increased liver size imply a rapid rate of fermentation and nutrient absorption. Put succinctly, these data illustrate a singl e point: giraffe are not cattle. Given the known differences between GR and CS, the likelihood of unmodified domestic ruminant feeding practices to maintain optimal health and nutritional status of captive CS appears questionable. However, the basic biologica l principles of ruminant digestion and metabolism discovered via domestic ruminant res earch may be able to facilitate improved nutrition for captive CS if vi ewed in light of CS rumi nants’ unique anatomy and physiology. In discussing the unique anatomical a rrangements of CS, Hofmann touches upon the need for further analysis of natural f oods in order to maintain captive CS under optimal conditions. He concludes by stating: “N eglect of original c onditions finally leads

PAGE 19

7 to a failing of the delicate anatomical, phys iological, biochemical-microbial balance of the concentrate selector’s ruminant stomach ” (Hofmann, 1973). The research presented in this thesis is an attemp t to further the understanding of captive CS nutritional status and requirements, and to examine possible lin ks between dietary factors and suspected nutritional pathologies, including PMS in captive giraffe.

PAGE 20

8 CHAPTER 2 REVIEW OF THE LITERATURE ON DIETARY PHYSICAL FORM AND CARBOHYDRATE PROFILE IN RUMINANT DIETS Physical Form Current feeding recommendations for z oo ungulates (Lintzenich and Ward, 1997) do not address two of the more recent areas of focus in ruminant nutrition research: dietary physical form and non-fiber carbohydrate (NFC) profile. The physical form of dietary components impacts the manner in whic h feedstuffs are processed in the digestive tract. Dietary physical form affects ma stication (Mertens, 19 97), saliva production (Allen, 1997), ruminal development (Behar ka et al., 1998), ruminal pH (Allen, 1997), rate and extent of ruminal fermentation (Mer tens, 1997), rate of di gesta passage (Allen, 1996), and the proportion of unfermented nutri ents passing into the lower GI tract (Firkins, 1997; Callison et al., 2001; Yang et al., 2001). Ma ny methods of quantifying the physical effectiveness of feeds by evaluating dietary part icle size, chewing behavior, or milk fat production have been proposed and used in domestic ruminant research. Physically effective fiber (peNDF) is one appr oach used to define the effectiveness of dietary particle size in maintaining rumina l (and animal) health and function (Mertens, 1997). The peNDF of a feed is defined as “the product of its neutral detergent fiber (NDF) concentration and its physical effectiv eness factor,” with physical effectiveness factor determined by the ability of the feed to promote a chewing response in the animal, as judged on a scale of 0 (not effective) to 10 (fully effective in promoting chewing). The

PAGE 21

9 physical effectiveness of a feed is, in essen ce, a function of partic le size and rate of particle size reduction. Effects of Particle Size on Mastic ation, Saliva Flow, and Ruminal pH Little reduction in feed particle size o ccurs once ingesta has passed from the rumen (Poppi et al., 1980). Chai et al. (1984) demonstrated th at initial mastication and rumination serve to reduce feed particle size. In steers fed long-stem alfalfa or brome hays, bolus content of particles > 3.35 mm was reduced 58 to 75% by initial mastication and 23 to 27% by ruminati on (Chai et al., 1984). The ability to stimulate chewing behavior appears to vary among forage types. Steers on high-concentrate diets spent more time ( P <0.10) chewing when fed wheat straw rather than alfalfa hay (Shain et al., 1999) The number of eating and ruminating chews per g of DM consumed were 2.04 and 3.41 by steers fed long-stem brome hay, but only 1.26 and 1.80 by steers fed long-stem alfalfa hay, which may have been attributable to differences in forage fragilit y, or rate of particle breakdow n during chewing (Chai et al., 1984). Dietary items with a larger physical size require more time to be consumed and generally have a greater ab ility to stimulate rumination. Long-stem or minimally chopped forages, with high NDF content a nd long particle length, have a greater stimulatory effect on mastication than do finely chopped forages, and a higher peNDF value than grains or pellets (Mertens, 1997). Eating and total chewing time in lactating Holstein cows offered a total mixed ration (TMR) increased ( P <0.05) with inclusion of additional alfalfa in the ration, and ruminati on and total chewing increased linearly with increasing particle length of alfalfa silage ( P <0.05) (Clark and Armentano, 2002). Time engaged in rumination decreased ( P <0.001) linearly as wheaten hay fed ad libitum to

PAGE 22

10 sheep and goats was progressively switched from chopped (1 cm) to ground (3.2 mm) and pelleted forms (McSweeney and Kennedy, 1992). Campbell et al. (1992) used ten Hereford steers to compare the ability of five diets to stimulate chewing: A – 80% pelleted concentrate, 20% long timothy hay (con trol); B – 80% pellet ed concentrate, 20% alfalfa cubes; C – 90% pelleted concentrate, 10% alfalfa cubes; D – completely pelleted ration using corn cobs as the primary NDF source; E – 80% coarse (unground grains) concentrate, 20% long timothy hay. Time e ngaged in rumination and eating behaviors are reported as minutes per gram of dry matter intake (DMI)/ BW0.75. Modifying the physical form of concentrates had no effect on eating ( P =0.702) or rumination ( P =0.954) times. Replacing timothy hay with alfalfa cubes decreased rumination from 2.58 to 1.38 to 1.47 ( P =0.001) but did not affect eating behavior ( P =0.897). Replacing the control ration with the all-pelleted diet d ecreased eating time from 3.55 to 2.98 ( P =0.063) and rumination from 2.58 to 1.29 ( P =0.001). Number of chews per gram of DMI/ BW0.75 also decreased on the completely pelleted ration, from 234 to 173 ( P =0.005) during eating, and from 162 to 76 ( P =0.001) during rumination. Saliva, with a pH of approximately 8.5 (C assida and Stokes, 1986), is a primary ruminal buffering agent, supplying approximate ly half of the bicarbonate entering the rumen of domestic cattle (Owens et al., 1998). The rate of saliva flow increases during periods of eating and ruminati on (Bailey, 1961). Therefore, decreasing time engaged in chewing behavior may decrease total daily saliva production per unit of feed consumed (Bailey, 1961; Maekawa et al., 2002). A study by Beauchemin et al. (2003) illustra tes the link between forage particle size, chewing activity, a nd ruminal pH. The effects of di etary particle size on lactating

PAGE 23

11 dairy cows were examined using a total mi xed ration (TMR) consis ting of 60% barleybased concentrate and 40% forage (DM basis). Forage consisted of alfalfa silage (AS) and alfalfa hay (AH) in a 50:50 or 25:75 rati o. Alfalfa hay was coarsely chopped (CH) or ground (GH) through a 4mm screen. Mean diet ary particle length (MPL) was highest for the TMR containing 50:50 AS:CH, followe d by 25:75 AS:CH, 50:50 AS:GH, and 25:75 AS:GH. Eating and rumination behaviors were recorded every 5 minutes for 24 hours. Rumination time per unit of DMI decreased with decreasing dietary pa rticle size (Table 2-1). The shortest rumination time was 4.6 hours/day or 13.5 minutes/kg of DMI. Total chewing time varied from 12.1 to 9.9 hours/day, and decreased with decreasing MPL. Ruminal pH was measured every 5 seconds fo r 48 hours by an industrial electrode placed in the ventral sac via ruminal cannulae. Ruminal pH was evaluated according to time above 6.2 and below 5.8, based on the observa tions of Russell and Wilson (1996) that ruminal microbial activity is compromised wh en pH falls below 6.2, and the premise that the incidence of sub-clinical acidosis increases when ruminal pH drops below 5.8. Cows fed 25:75 AS:AH had a greater mean ruminal pH ( P =0.10) and time above pH 6.2 ( P =0.08) than cows fed 50:50 AS:AH, perhaps due to decreased DMI ( P <0.10) and increased time spent eating ( P =0.01). Ruminal pH status was mainly affected by forage particle size, and was improved by feedi ng chopped, rather than ground, alfalfa hay (Table 2-1). Ruminal pH is influenced by the relative co ncentrations of acid s, bases and buffers present at any given time. Fermentation of NFC found in high concentrations in concentrates and cereal grains results in ra pid production of organi c acids, while peNDF consumption stimulates saliva flow. C oncurrent consumption of NFC and peNDF

PAGE 24

12 sources can be facilitated by combining diet components in a TMR rather than offering forage and concentrate separately. Maek awa et al. (2002) us ed eight ruminally cannulated lactating Holstein cows arrange d in a double 4 x 4 Latin square design to examine the effects of offering whole crop barley silage and steam-rolled, ground barley grain-based concentrate on chewing activitie s, saliva production and ruminal pH. Four diets were offered: separate in gredients (SI) offered at a fora ge to concentrate (F:C) ratio of 50:50 (DM basis), and ad libitum TMR with F:C ratios of 60:40, 50:50, and 40:60. When SI were fed, silage was offered ad libitum and grain was offered at 50% of previous DM consumption to facilitate consumption of a 50:50 F:C ratio. However, animals fed SI elected to consume an actual F: C ratio of 43:57, illustrating that when diet components are fed separately, the animals ma y choose to consume a different F:C ratio than the formulated ration. Cows were fed twice daily, with silage fed 1 hour after the concentrate on SI treatment. Eating and ru minating activities were recorded every 5 minutes for 24 hours. Ruminal pH was measur ed every 5 seconds for 24 hours, and mean pH for each 15-minute period was recorded. Saliva samples were collected at the cardia during feeding and resting. Salivation coul d not be measured during rumination, so salivation rate was assumed to be th e same as that observed during eating. Salivary secretion rate was 2.2 times great er during eating than resting. Linear effects of increasing silage c oncentrations in the TMR showed only numerical increase in minutes/ day spent eating ( P =0.18), but increased minutes/ day spent ruminating ( P =0.03), increasing to tal chewing time ( P =0.01) from 741 to 757 and 848 minutes/day on F:C ratios of 40:60, 50:50, and 60:40, respec tively. Total chewing time by cows fed the SI diet (736 minutes/ day) was similar to that of cows fed the 40:60 TMR. However,

PAGE 25

13 SI was consumed at a faster ( P =0.02) rate (10.9 minutes/ kg of DM) than the 40:60 TMR (13.5 minutes/ kg of DM). Concentrates required less mastication before being swallowed, and were therefore consumed more rapidly than silage or TMR ( P <0.01), decreasing the amount of saliva produced pe r unit of feed (ml/g of DM) from 4.43 on silage to 1.19 on concentrates ( P <0.01). The main diet effect on ruminal pH was not significant, but numerical differe nces presented in Table 2-2 s uggest that diet did alter pH characteristics. The postprandial decline in ruminal pH was greatest for cows fed SI, and mean ruminal pH in SI cows was below the 5.8 benchmark suggested for increased risk of subacute ruminal acidosis. Feeding TMR ra ther than SI appeared to decrease risk of ruminal acidosis by preventing higher than in tended consumption of concentrates and increasing saliva production via increased mastica tion at the time that rapidly fermentable concentrates were consumed. Effects of Particle Size on Digest a Passage, Intake, and Fermentation Kennedy et al. (1992) reported a linear increase in DMI ( P <0.001) and linear decreases in ruminal DM retention time ( P =0.001) and DM digestibility ( P <0.01) when wheaten hay fed ad libitum to sheep and goats was progressively switched from hay chopped to 1 cm (c), to a 2:1 ratio of c + hay ground to pass through a 3-mm screen and pelleted (p), to a 1:2 ratio of c+p, to an a ll p diet. Dietary particle size can influence voluntary intake, rate of digesta passage, and rate and extent of ru minal fermentation. Since the rumen is a dynamic system, multiple interactions, exceptions, and caveats must be considered. The following includes a brief, and by no means comprehensive, discussion of such considerations. Although multiple factors, including hydrati on and density, influence the rate at which feed particles leave the rumen, resist ance to outflow increases with increasing

PAGE 26

14 particle size (Poppi et al., 1980). Poppi et al. (1980) suggested a high resistance for particles greater than 1.18 mm, since less than 5% of particle s leaving the rumen of sheep fed chopped (2 cm) hay were retained on a 1.18 mm mesh sieve. Small particles may flow from the rumen more rapidly than la rge particles (Weston and Cantle, 1984). Long forage particles form a mat in the rumen, whic h can trap small particles, retaining them for a greater length of time (Grant, 1997). As a result, decreasing forage particle size may decrease overall ruminal DM re tention (Bernard et al., 2000). An increased rate of digesta passage can increase voluntary intake by reducing the constraints of rumen fill. A direct relations hip has been demonstrated between increased ruminal contents and decreased voluntary in take (Campling and Balch, 1961); Schettini et al., 1999). When long forages are c onsumed, rumen distention resulting from restricted digesta flow may limit voluntar y intake (Allen, 1996), which may prevent intake of sufficient energy to meet anim al requirements (Miller and O'Dell, 1969). Conversely, fine particle size in an entire ration, particularly in high concentrate rations, can also reduce voluntary intake (Krause a nd Combs, 2003), likely due to unfavorable alterations in rumen conditions, such as a decline in pH (Nocek, 1997). Dietary particle size impacts ruminal digestibility in part by influencing the balance between the length of time f eed components are retained in the rumen and the rate at which they are fermented. Ruminal fermenta tion of dietary nutrients takes time. An increased rate of particulate outflow from the rumen decreases the amount of time feed components are available for microbial fermen tation. Ruminal digestibility of dietary components, particularly of slowly fermenti ng fiber, may decrease as a result (Pasha et al., 1994). Conversely, chopping or grinding feed components increases the surface area

PAGE 27

15 available for microbial a ccess (Owens and Goetsch, 1988) increasing the rate of fermentation. As a result, decreasing the par ticle size of a particular feed component may increase the extent of its digestion in th e rumen (Callison et al., 2001). However, when rapid fermentation is not balanced by incr eased buffers and disappearance of organic acids from the rumen, ruminal pH will decreas e, the microbial populat ion will be altered, and nutrient digestibility may d ecrease (Russell and Wilson, 1996) Changes in dietary physical form alter the ruminal environment, and may result in a shift in VFA profile. In a study by Krause et al. (2002), four TMR rations with forage to concentrate ratios of 39:61 (DM basis) were used to investigate th e effects of forage particle size and the concentr ation of dietary ruminally fe rmentable carbohydrates (RFC) in diets of equal NDF content. Diets were offered ad libitum to eight ruminally cannulated, lactating Holstein cows. Dietary concentrates were cracked-shelled corn (low RFC) or ground high-moisture shelled corn (high RFC). Dietary forages were chopped alfalfa silage of mean particle length of 13.6 (coarse) or 3.7mm (fine). Decreasing forage particle size decreased the amount of time engaged in chewing behavior (Table 2-3). Both decreasing fora ge particle size and increasing RFC decreased mean ruminal pH, increased time (hours per day) and area (time*pH units/day) of ruminal fluid below pH 5.8, and decreased the ratio of acetate to propionate by increasing ruminal propionate concentration (Table 2-3) The percentage of dietary particles retained on the top screen of a Penn State particle separator was positively correlated (0.61) with minutes of chewing per day ( P= 0.0003) and, to a lesser extent, negatively correlated (-0.32) with time that ruminal fluid pH was <5.8 ( P= 0.09).

PAGE 28

16 Jorgenson and Schultz (1963) fed lactat ing cattle 7.26 kg of ground corn daily, along with long-stem (control) or pelleted alfalfa hay in ad libitum amounts. Feeding pelleted hay increased DMI from 1.1 to 1.2 kg/ 45 kg of body weight ( P <0.05), increased total ruminal VFA concentration from 600 to 844 mg/ 100 ml of ruminal fluid ( P <0.05), and altered the VFA profile of ruminal fluid. As a percentage of total VFA, acetate decreased from 60.3 to 55.6% ( P <0.05), propionate increased from 20.1 to 27.0% ( P <0.05), and butyrate decreased numerically from 17.0 to 15.3% ( P >0.05). A second trial compared the same control diet with an experimental diet of 16.3 kg of a 50:50 mixture of alfalfa: corn pellets, plus ad libitum alfalfa pellets. Changes in VFA profile of the ruminal fluid were similar to the first tr ial. Intake, however, declined from 1.22 to 1.18 kg/ 45 kg of body weight ( P <0.05) due to difficulty in keeping the cows consuming the pelleted feed. Blood glucose concen trations increased from 49.1 to 57.4 mg% ( P <0.05) in cows fed the all-pe llet diet. Although ruminal pH was not reported in this study, subacute ruminal acidosis (SARA) may be caused by an elevation in total VFA (Stone, 2004), and can resu lt in decreased or variable intake (Nocek, 1997). In diets for domestic ruminants, forage ha s long been the staple physically effective component, and the effects of forage part icle length on digestion parameters and production performance have been investigated widely. In more recent years, byproduct (non-forage) feeds have come into use as eff ective fiber sources in the livestock industry. The peNDF value of nonforage fiber sources is considerably lower than long-stem forages, but may be higher than some form s of concentrates, grains, and ground forages (Mertens, 1997).

PAGE 29

17 Partial replacement of forage with co ttonseed hulls (CSH) has been shown to increase DMI and decrease NDF and DM digestib ility (Moore et al., 1990; Theurer et al., 1999) in cattle, suggesting an a ssociated increase in passage ra te. Dry matter intake as a percent of body weight in steers fed sorghum grain and chopped (15 cm screen) alfalfa hay increased ( P <0.05) when half of the alfalfa wa s replaced with CSH or chopped (2.5 cm screen) wheat straw (Theurer et al., 1999). However, DMI intake per kg of body weight gain was lower for steers fed alfalfa, suggesting an increased efficiency of feed utilization with alfalfa as compared to CSH. Further effects were noted when CSH re placed long (theoretical 22.3 mm) or short (theoretical 4.8 mm) cut corn silage in TMR offered to lactating Holstein cows (Kononoff and Heinrichs, 2003). The inclusion of CSH at 8% of dietary DM reduced the concentration of corn silage from 57.4 to 45.8% of dietary DM. Physically effective NDF, estimated according to (Mertens, 1997), did not differ across treatments, and increased sorting behavior was noted when co ws were fed long corn silage. It is therefore not surprising that reducing corn silage part icle size did not affect ( P >0.05) chewing activity, DMI, ruminal pH, or appa rent total tract digestibility of total nonstructural carbohydrates, NDF or acid detergent fiber (ADF). Total chewing time per kilogram of NDF intake tended ( P <0.10) to increase by cow fed diets containing long corn silage. The inclusion of CSH tended ( P =0.06) to decrease to tal chewing time per kilogram of DM consumed, and decreas ed mean ruminal pH from 6.24 to < 6.17 ( P =0.05). Cottonseed hulls increased DM and NDF intake ( P <0.01) but did not affect DM digestibility.

PAGE 30

18 Potential Implications of peNDF for Captive Giraffe From 1974 to 1978, a series of digestion tr ials comparing grass and alfalfa hays were conducted using thirty herbivore speci es at five zoos (Foose, 1982). For each dietary treatment, 14 days were allowed for acclimation to diets, followed by 10 days of intake measurements. Total fecal producti on was collected during the final 4 days. Individual animal intake was recorded, and orts were composited by species and treatment (grass or alfalfa hay) for chemical analysis. Daily fecal output of each animal was weighed, and pooled subsamples from each of the 4 days were composited for analysis. The report mentions that “in most cases, it was possible to collect (fecal) accumulations only at 24 hour intervals.” Gi ven that inability to collect samples soon after defecation resulted in sample trampling and contamination that prevented total fecal collection in one giraffe study (C lauss et al., 2001), giraffe digestibility results from the Foose (1982) study may be inaccurate. Intake and nutrient extracti on results are reported for thre e giraffe fed alfalfa hay and one giraffe fed timothy grass hay (Table 24). Foose (1982) reports that grass hay digestion trials on several giraffe had to be discontinued because the animals were “conspicuously languishing on the diets.” Gira ffe used were 2 and 3 years old, and thus were likely to still be growing. Body weight could not be measured, and was estimated from the literature. Organic matter intake (% of BW) and digestion coefficient ((amount ingested – amount defecated) / amount i ngested) x 100% were 0.45 and 57.11 for grass hay, and 0.89 and 60.70 for alfalfa hay. Of th e 28 ungulate species in this study, giraffe had the lowest intake as a percentage of body weight, suggesting that the ruminal fill effects of long forage may have an especially high impact in giraffe.

PAGE 31

19 Feeding all-forage diets to high produci ng dairy cows can prevent adequate energy intake, but feeding high-concentrate rations ha s potential to cause a variety of metabolic and health problems (Miller and O'Dell, 1969). It has been suggested (Clauss et al., 2002b) that captive giraffe fed a traditional hay/ concentrate diet face a nutritional dilemma. Giraffe consuming a high proportion of hay will increase ruminal fill and decrease intake. Those that consume a high pr oportion of concentrates will increase their potential risk of ruminal acidosis. The appropriate particle size for main taining optimal diet utilization and gastrointestinal health in CS remains to be determined, but it has been suggested that giraffe are ill-suited to consumption of longstem forages and the formation of a ruminal mat similar to that in domestic ruminants (Cla uss et al., 2002b). The giraffeÂ’s diet in the wild consists primarily of polygonal leaves as opposed to the elongated grasses and hays consumed by GR (Clauss et al., 2002b). Thes e physical as well as chemical features including density factors may explain why stratification of ruminal contents occurs in wild GR, but not in wild CS ( Hofma nn, 1973; Clauss et al., 2001). Anatomical differences between GR and CS (Hofmann, 1973) suggest a naturally faster passage rate of digesta in CS ruminants than in GR ru minants, which has been observed in captive giraffe (Hatt et al., 1998). Giraffe appear to be adapted to rapid rates of fermentation and nutrient absorption (Hofmann, 1973). However, feeding high-concentrate, low-fora ge diets is unlikely to enhance captive giraffe nutrition. It seems lik ely that even in giraffe, aci d production from ruminallydegraded organic matter (particularly rapidly fermenting NFC) must be balanced with the ruminal dilution and salivary stimulation eff ects of an appropriate form of peNDF (Allen,

PAGE 32

20 1997). Saliva production in wild giraffe may be high due to the oral stimulation and time involved in selective feeding pa tterns. One of the challenges in feeding captive giraffe is providing diets with a physical form that will stimulate mas tication and saliva flow to maintain a balanced rumen pH, while avoidi ng an unnatural reduction in rate of passage and dietary intake. Dietary Carbohydrate Profile Carbohydrate Fractions Dietary carbohydrates can be divided into two basic frac tions: fiber and nonfiber carbohydrates (NFC). The chemical bonds betw een sugar residues in dietary fiber cannot be broken by mammalian enzymes. As a re sult, fiber cannot be digested by mammals themselves, but may be utilized by microbes in the digestive tract. The end products of microbial fermentation can then be abso rbed and utilized by the host animal. Fermentation of cellulose and hemicelluloses is typically slow (2 to 14% digestion/ hour) (Sniffen et al., 1992). Nonfiber carbohydr ates, on the other hand, are generally associated with rapid rates of ruminal fermenta tion, with digestion rate constants of 75 to 400% digestion/ hour for sugars, and 5 to 50% digestion/ hour for starch and pectin (Sniffen et al., 1992). With the exception of neutral detergent-soluble fiber (NDSF), NFCs may be digested by mammalian enzyme s. The NFC may be further divided into sugars (mono-, di-, and oligosaccharides), st arch, organic acids, a nd NDSF (Hall et al., 1999). As we will see, mounting evidence sugg ests that these NFC fractions differ in their fermentation characteristics. As the science of carbohydrate analysis has progressed, new techniques have allowed evaluation of carbohydrates in increa sing detail. Use of different techniques over time can make the literature difficult to interpret. For the sake of clarity,

PAGE 33

21 carbohydrate fractions found in the different anal ytical systems discussed in this section are listed below. Proximate analysis system Crude fiber (CF): cellulose, acidand alkali-i nsoluble hemicellulose, acidand alkali-insoluble lignin Nitrogen-free extract (NFE): sugars, starch, pectic substances, organic acids, fructans, acidand alkali-soluble hemice llulose, acidand alkali-soluble lignin Under the proximate analysis system, fiber (hemicellulose) and li gnin can appear in either the CF or NFE fraction. The detergen t system, however, dis tinguishes between the fiber and non-fiber carbohydr ate fraction of feeds. Detergent system Neutral detergent fiber (NDF): cellulose, hemicellulose, lignin Acid detergent fiber (ADF): cellulose, lignin Nonfiber carbohydrates (NFC): sugars, starch, pectic substances, organic acids, fructans, and other carbohydrates soluble in neut ral detergent Carbohydrates in Natural CS Diets The methodology for differen tiation of NFC into sugars, starch and NDSF has only recently been applied to animal feedstuffs. (Dierenfeld et al., 2002) examined sugar and starch concentrations in 8 fr uit and 2 flower species consumed by wild duikers in the Democratic Republic of Congo. Samples contai ned 2 to15 times more sugar than starch, with DM percentages of sugars at 0.16 in one sample and 3.19 to 15.71 in 9 samples, and DM starch percentages of 7.43 in 1 sample, 0.37 to 1.96 in four samples, and less than 0.1 in 5 samples. Duikers are considered frugivores (fruit eaters), consuming large

PAGE 34

22 proportions of fruits, flowers a nd seeds. Hence, no leaf material was included in this report. Unlike duikers, giraffe are considered folivor es (foliage eaters). Plants consumed by wild giraffe vary widely with seas on and geographic location. While Acacia spp appear to be the most commonly reported di etary item ( Foster, 1966; Dagg and Foster, 1976; Furstenburg and Van Hoven, 1994; Ciof olo and LePendu, 2002; Caister et al., 2003), giraffe in a given location may consum e as many as 66 plant species over the course of a year (Leuthold and Leuthold, 1972). Trees and sh rubs comprise the bulk of the diet, with limited vine and herb consum ption (Leuthold and Leuthold, 1972). Grass consumption appears either non-existent (L euthold and Leuthold, 1972) or negligible (Ciofolo and LePendu, 2002). Field and Ro ss (1976) observed that woody plants comprised > 93% of the diet, and “grass appeared to be eaten by accident when enmeshed with other food.” Plant portions consumed ar e primarily leaves and stems, but may also include fruits, flowers and bark (Leuthold and Leuthold, 1972; Caister et al., 2003). The diet of giraffe in Niger was reportedly co mposed of 86% leaves, 8.5% stems and 5.5% flowers and fruits. However, during the dr y season (December to April), the diet was composed of 45% leaves and stems, 44% fr uits and 11% flowers (Ciofolo and LePendu, 2002). Wild CS are generally thought to consume a “rich, rapidly fermenting diet” (Hofmann, 1973). Pellew (1984) reported mean daily nutrient intakes of 2.22 to 3.12 kg of crude protein (CP), 0.53 to 0.97 kg of et her extract (EE), 4.33 to 8.63 kg of acid detergent fiber (ADF), and 5.48 to 7.73 kg of nitrogen-free extract (N FE) for giraffe in the Serengeti. Composition of leaves (DM basis) from 10 plant species consumed by

PAGE 35

23 giraffe in Niger during the dry season ra nged from 8.2 to 28.6% CP, 0.8 to 6.8% crude fat, 1.55 to 16.2% crude fiber (CF), and 19.8 to 71.93% NFE (Caister et al., 2003). Mean composition of browse leaves (DM basis) from the Narus Valley (Uganda) for the months of January, March, April, May a nd December ranged from 11.51 to 22.37% CP, 2.36 to 3.01% EE, 19.33 to 33.48% CF, and 36.80 to 50.32% NFE. Browse stems contained 5.78 to 9.35% CP, 1.17 to 1.62% EE, 32.49 to 48.80% CF, and 35.91 to 52.18% NFE (Field and Ross, 1976). The high NFE content in these studies seems to suggest a high concentration of nonfiber car bohydrates. However, it is important to remember that NFE contains not only sugars, starch and soluble fiber, but also variable amounts of hemicellulose and lignin, and ther efore is not a true representation of NFC content. In our search of the literature, we fo und no published data on NFC fractions in foliage. Analysis of leaves collected in Oc tober from three tree, one shrub, and one grass species used for zoo animal enrichment s howed higher concentra tions (DM basis) of sugars (6.5 to 13.2) than starch (0.3 to 3.1) in four of five species (Table 2-5; R. Ball, personal communications). Acacia leaves contained 37.2% NFC, 6.5% sugars, and 0.5% starch, suggesting that NDSF and organic ac ids may account for approximately 81% of the NFC fraction. As will be discussed belo w, the differing NFC fr actions do not digest in the same manner in a fermentative environment. In order to improve our understanding of their nutritional require ments, and optimize nutrition of captive folivorous CS species, NFC analysis of natu ral dietary foliage needs to be further explored.

PAGE 36

24 Effects of NFC Source on Fermentation Characteristics and Animal Performance Domestic ruminant feeding st udies suggest that higha nd lowstarch NFC sources differ in digestion characteristics and fermen tation end products. Grains such as corn, oats, wheat, and barley contain starch as the main nutrient component (Huntington, 1997). Supplemental energy sources having less starch include molasses and sucrose as sugar sources, as well as b eet pulp and citrus pulp, whic h have substantial NDSF and sugar contents. Carbohydrate feeding trials that have i nvolved addition or subtraction of whole feeds generally alter multiple ration characteristics, ra ther than carbohydrate profile alone. This may explain the variability in DMI and body weight changes in livestock fed diets in which NDSF or sugar sources repla ced fibrous or starchy feeds. Increasing supplemental NDSF sources (generally beet pul p or citrus pulp) wh ile decreasing starchy feeds (generally barley or corn) has both in creased DMI in cattle (C hester-Jones et al., 1991) and had no effect on DMI in sheep (B en-Ghedalia et al., 1989). Supplemental NDSF tended ( P = 0.05) to decrease average daily gain in one study (Chester-Jones et al., 1991), but showed no effect on body weight in others (Bhattacharya and Lubbadah, 1971; Friggens et al., 1995). Altering dietary NFC profile has changed fermentation patterns, ruminal pH, and diet digestibility. Switching la ctating cows fed a basal diet of freshly cut grass (85% Lolium perenne ) from a supplement (36% of dietar y DM) containing 47.5% corn meal and 50% hominy to a supplement containing 82.5% sugar beet pulp and 15% soybean hulls increased ( P <0.05) ruminal OM and NDF diges tibility and VFA concentrations (Van Vuuren et al., 1993). Ben-Ghedalia et al. (1989) used four ruminally and duodenally cannulated Merino rams to compare digestion characterist ics of supplements

PAGE 37

25 of dried citrus pulp (84.4% of supplement DM) with supplement containing citrus pulp (20.4%) and barley (76.5%). Isonitrogenous supplements DC P (84.4% citrus pulp) and B (20.4% citrus pulp, 76.5% barley) were offered with lucerne hay in an 80:20 supplement: forage ratio. Sheep consuming DCP had grea ter ruminal pH (6.42) than those consuming B (6.18). Total tract apparent digestibilit y of OM did not differ between diets, but apparent NDF digestibility was greater by sheep fed DCP (79.4%) than those fed B (63.6%). The results indicate that the ferm entation of DCP created a more favorable ruminal environment for fermentation of NDF. In vitro investigation of microbial CP yield, detected as tr ichloroacetic acidprecipitated crude protein (T CACP), also suggested differences in NFC fermentation patterns (Hall and Herejk, 2001). When is olated NDF from bermudagrass hay was incubated with sucrose, corn starch, or citr us pectin in media containing mixed ruminal microbes, maximum TCACP synthesis (mg/g s ubstrate OM) was greate r for starch (85.6) than sucrose (73.3) or pectin (75.4) ( P <0.05). Temporal patterns of TCACP yield differed as well. Sucrose showed no detectab le lag phase, with greatest yield of TCACP at the 12-hour sampling, and declined only gradually over time. The TCACP reached peak yield at 12 and 16 hours during fermen tation of substrate containing pectin, followed by a rapid decline. When starch was fermented, peak TCACP yield was not reached until 16 hours, also followed by a rapid decline. Bhattacharya and Lubbadah (1971) used cattle and sheep to examine the effects of replacing corn with beet pul p in high (approximately 75%) concentrate diets. Three experiments were conducted, with the same four supplements fed in each experiment. In all three experiments, the c ontrol supplement (Die t I) consisted of corn (73%), soybean

PAGE 38

26 meal (21%), tallow (0.8%), bonemeal (2.7%), limestone (2.0%) and salt (0.5%). Beet pulp was used to replace 0, 50, 75, and 100% of the corn in Diets I, II, III, and IV, respectively. As beet pulp concentrations increased, concentrations of tallow and bonemeal were increased, and limestone decreas ed, to maintain similar levels of fat, phosphorus, and calcium among diets. Experiment I used four lactating Holstein -Friesian cows in a 4 x 4 Latin square design, with 4-week periods, to evaluate di etary effects on body weight gain, milk yield, and milk fat. Animals were offered 19 kg of concentrate and 5 kg of alfalfa hay daily. No significant differences were observed. It was noted that bloat symptoms occasionally seen during pre-experimental a nd experimental periods in co ws on the control diet were not observed in cows offered the beet pulp diets. Experiment II consisted of two 17-day trials using eight wethers housed in metabolism crates, with two animals randomly allotted to each feeding treatment during each trial. Alfalfa hay in this experiment was ground (size unreported) and mixed with concentrate in a 26.2: 73.8 ratio. Diets were offered twice daily for three hours each feeding. Sample collection in each trial incl uded total feces and urin e (final 7 days) and jugular blood samples at three hours postfeeding on the day following the end of the collection period. Apparent to tal tract DM digestibility was not affected by diet. Replacement of Diet I (73% corn supplement) with Diet IV (73% beet pulp supplement) decreased the apparent percent digestibil ity of CP (78.2 to 72.5%) and NFE (87.1 to 85.5%) and increased the apparent percent digestibility of CF (46.1 to 72.4%) ( P < 0.01). No dietary effect on blood glucose or VFA concentration was reported.

PAGE 39

27 In Experiment III, rations “similar to those in Experiment 2” were offered ad libitum to four ruminally fistulated Holstein steer s. Beginning with the control ration and progressing to higher beet pulp concentrati ons, animals received each ration for seven days, with four days gradual adjustment betw een diets. Intake (a verage 7.65 kg/ day) did not differ across diets. Rumi nal fluid measures at the end of each period showed differences ( P <0.01) in VFA concentration (mmole/ L) response among all treatments, and increased VFA concentrations in animal s on Diets II (138.7), III (157.5) and IV (141.3) as compared to control animals (113.3 ). Ruminal pH was numerically lower on Diet IV (5.9) than on Diets I, II, and III (6.2, 6.3, and 6.2 respectively), and lactic acid concentration was not affected. The author s observed that when animals received the control diet, “the rumen was full of gas and frothy foam. As the proportion of … beet pulp increased…, gas and foam gradually disa ppeared, and the app earance of the rumen ingesta resembled that under hay-feeding conditions.” Notable treatment differences in fermen tation characteristics were observed in ruminally cannulated Angus x Hereford steers (n=20) offered low-qual ity tallgrass-prairie hay (chopped through a 75-mm screen, offere d at 130% of previous intake) only (control), and hay supplemented with starch, glucose, fructose or sucrose at 0.30% of BW/ day (Heldt et al., 1999). Supplemented steers also received ruminally degradable protein (RDP) (sodium caseinate, 91.6% CP ) at 0.031% and 0.122% of BW/ day in Experiments 1 and 2, respectively. Supplements were administered intraruminally. In Experiment 2, total tract OM digestibility was 7.9% greater in animals supplemented with sugars as compared to those receiving starch ( P =0.04). In agreement with Ben-Ghedalia et al. (1989), ruminal pH and total tract NDF digestibility were lower for animals

PAGE 40

28 supplemented with starch than fo r those supplemented with sugar ( P < 0.05). Volatile fatty acid concentrations differed between control and supplemented animals, and between animals supplemented with sugar or st arch (Table 2-6). Ruminal concentrations of butyrate and lactate were greater in animal s supplemented with sugars than in control animals or animals supplemented with starch ( P < 0.01). In both experiments, timerelated fermentation patterns were dramatically different for sugars and starch. With all three sugar treatments, pH declined rapidl y, reaching the lowest point at 3 hours after supplementation, followed by a rapid recovery. Decline in pH was slower for the starch treatment, reaching the lowest poi nt 9 hours after supplementation. Interaction Between Dietary Components Heldt et al. (1999) also illu strates the possible impact of other dietary components, in this case protein, on fermentation characteri stics of NFC fractions. In Experiment 1, where RDP was fed at 0.031% of BW/ day, OM and NDF digestibility did not differ between supplement treatments, and NFC supplementation depressed NDF digestibility when compared to the hay-only control. Increasing RDP from 0.031 to 0.122% BW/ day (Experiment 2) resulted in higher OM and NDF digestibilities in animals receiving supplement vs. control, and sugars vs. star ch. Carbohydrate treatment affected mean ruminal pH at high, but not low, RDP amount s. Other factors beyond RDP can alter the effect of dietary NFC. The physical form of the NFC source can affect the site of digestion and the extent of ruminal digest ion (Callison et al., 2001) The proportion of dietary forage can affect ruminal VFA pr oduction response to i ndividual NFC feeds (Friggens et al., 1998). Dietar y characteristics other than NFC profile alone may affect animal response to NFC supplementation.

PAGE 41

29 Potential Implications of Dietary NFC Profile for Captive Giraffe Clemens et al. (1983) reported the molar pr oportions of VFA in the digestive tracts of 16 species (4 CS, 5 intermediate feed ers (IM), and 7 GR) of free-ranging African ruminants, using animals sacrificed during wildlife management programs. The digestive tract was separated by ligatures into six se gments, and representative content samples from each segment were strained through chees ecloth. Supernatant was acidified with concentrated H2SO4 and refrigerated for later analysis. Sample collection and field analysis were generally completed within 1 hour after death. Of all species, giraffe had the highest ruminal proportion of acetate (73.2 + 1.6 molar %; 60.2 to 72.9 molar % in other species), and the second lowest propor tion of propionate (14.1 + 0.5 molar %; 12.8 to 22.8 molar % in other species) in ruminal fl uid. The acetate: propionate ratio (5.21 + 0.13) in ruminal fluid from giraffe was exceeded only by the eland (5.71 + 1.38) and followed closely by the oryx (5.13 + 0.11), with other species ranging from 2.92 to 4.65. Traditionally, high acetate c oncentration is associated with fiber fermentation, and elevated propionate concentra tion is associated with NFC fe rmentation. Since giraffe are thought to consume diets relatively rich in NFC and low in NDF, with extensive fiber fermentation impeded by a rapid rate of di gesta passage from the rumen, high acetate/ low propionate concentrations in the wild gi raffe rumen appear paradoxical. However, domestic ruminant studies reveal that the various NFC fractions differ in fermentation and VFA production characteristics. A lthough VFA disappearance rates and the unknown nutrient content of feedstuffs consum ed prior to sample collection cannot be discounted as possible causes for the findings of Clemens et al. ( 1983), another potential contributing factor lies in the still unquantif ied NFC fractions of giraffe diets consumed in the wild.

PAGE 42

30 In domestic sheep and cattle, ruminal acet ate and propionate profiles have been altered by modification of dietary NFC. Wh en high amounts of citrus pulp (84.4% of dietary DM) were fed to sheep in place of a 20.4% citrus pulp and 76.5% barley diet, propionate decreased numerically from 17.6 to 14.4 molar % (P>0.05) and acetate increased from 65.0 to 69.1 molar % (P<0.05) (Ben-Ghedalia et al., 1989). Using continuous culture in vitro fermentations with mixed ruminal microbes from cattle inoculum, Ariza et al. (2001) observed changes in molar con centrations of propionate and acetate when starch was decreased from 24.0 to 11.0% and NDSF was increased from 8.8 to 14.4% by altering the substrate ratio of hominy feed to citrus pulp. Propionate decreased from 22.7 to 16.7 molar % and acet ate increased from 62.6 to 68.9 molar % (P<0.04). The acetate: propionate ratio increas ed from 2.8 to 4.1 (P=0.01). Strobel and Russell (1986) provided six carbohydrate substr ates (starch, sucrose, cellobiose, xylan, pectin, and a mix of the preceding in equal pa rts) to mixed ruminal bacteria at 1 mM/ h for 10 h at an initial pH of 6.7 (neutral) or 6.0 (low). Propionate c oncentrations across treatments were numerically lowest for pec tin, though the differences were not significant (P>0.05) at low pH. Less (P<0.05) propionate was produced in response to pectin (1.3 mM) than starch (2.9 mM) or cellobiose (2.7 mM) at pH 6.7. At neutral pH, millimolar concentrations of acetate from fermentati on of pectin, cellobiose, starch, mixed carbohydrates, sucrose, and xylan were 10.1, 6.4, 5.1, 4.8, 4.7, and 3.6 respectively, and were greatest (P<0.05) from pectin fermen tation. At low pH, pectin fermentation resulted in greater (P<0.05) acetate concentrations than fermentation of other singlecarbohydrate treatments, and numerically gr eater (P>0.05) acetate than from mixed carbohydrates. Acetate concentrat ions from pectin fermentation at low pH were half (5.0

PAGE 43

31 mM) those observed at neutral pH (10.1). It should also be noted that decreasing pH increased (P<0.05) lactate production from starch (0.9 to 4.1 mM), sucrose (3.7 to 8.3 mM), and cellobiose (2.3 to 8.2 mM) wh ereas lactate was not detectable from fermentation of pectin or xylan at either pH The NFC profile of giraffe diets consumed in the wild has not been determined, but hi gh concentrations of pectic substances in native feeds could explain the ruminal acetate and propionate concentrations observed in free-ranging giraffe. The extraordinarily high ruminal papillati on in wild giraffe results in an average 24x increase in ruminal surface area (Hof mann, 1973), denoted as surface enlargement factor (SEF; (papillary surface + base surf ace)/ base surface, wher e papillary surface = length x mid-level width of papillae x 2 (Hof mann et al., 1988)). By contrast, mean ruminal SEF in two captive giraffe was 1.90 a nd 2.67, owing to a dramatic decrease in papillae size and density (Hofmann and Mate rn, 1988). The diet of the two captive giraffe was ad libitum lucerne hay, browse (a mount not reported), variable amounts of produce, and a concentrate mixture consisti ng primarily of oats, barley, germinated wheat, soya, lucerne pellets, and maize. Butyrate is thought to be the most influe ntial VFA in ruminal papillae development (Van Soest, 1994). Changes in ruminal butyrat e concentrations have been variable when sugar or pectin sources were fermented in co mbination with varying proportions of starch or peNDF. It is interesting to note that in both NFC e xperiments conducted by Heldt et al. (1999), ruminal concentrati ons of butyrate were lower than propionate for animals consuming control and starch-containing treat ments, but greater in animals consuming diets supplemented with sugar, due to an in crease in butyrate concentrations. Strobel and

PAGE 44

32 Russell (1986) also observe d greater (P<0.05) in vitr o butyrate production from fermentation of sucrose as compared to star ch at neutral (6.7) pH but no difference at low (6.0) pH. These observations raise ques tions about a potentia l relationship between high ruminal SEF in wild giraffe and VFA production from fermentation of the unknown NFC fractions of foliage compared with low SEF in captive giraffe offered high-starch concentrates. Table 2-1. Effects of particle size of alfa lfa-based dairy cow di ets on chewing activity and ruminal pH. ---------------Dietsa--------------AS:AH 50:50 AS:AH 25:75 Item CH GH CH GH SE Pc MPLb, mm 9.78 5.137.50 4.42 0.56 0.01 Eating activity (min/kg DM) 12.5 10.8 14.8 15.5 1.7 >0.15 Rumination time (min/kg DM) 19.1 15.7 20.8 13.5 1.6 0.01 Total chewing time (min/kg DM) 31.6 26.5 35.6 28.9 2.7 0.01 Mean rumen pH 5.97 5.786.18 5.90 0.15 0.02 pH > 6.20, hours 7.3 3.5 11.0 7.6 3.3 0.11 pH < 5.80, hours 7.5 13.0 5.1 11.7 2.6 0.01 pH > 6.20, area (pH*hours) 6.9 10.5 4.4 10.2 2.2 0.01 pH < 5.80, area (pH*hours) 2.1 3.7 0.9 4.5 1.1 0.01 aAS:AH = Ratio of alfalfa silage to alfalfa hay; CH = chopped hay; GH = ground hay; overall diet was 60% barley-based concentrate and 40% forage on a DM basis. bMPL = Mean particle length. Fifty percent of diet ary particles are greater or less than this length. cP -value for forage particle size effect. (Beauchemin et al., 2003) Table 2-2. Effects of con centrate level and feeding ma nagement on ruminal pH of lactating dairy cows. -----------------Diets-----------------SIa ---Total Mixed Ration--Item 43:57b 40:60b 50:50b 60:40b SE Pc Mean pH 5.77 5.84 5.99 5.98 0.08 0.25 Minimum pH 5.14 5.26 5.40 5.30 0.06 0.06 Percent of time under pH 5.8 55.36 46.25 33.25 34.78 8.0 0.26 Area under pH 5.8 (pH*hours/ day) 4.99 3.52 2.60 2.37 0.81 0.19 aSeparate ingredient diet. bForage-to-concentrate ratio consumed. cP -value for main effect of diet. Linear and quadratic contrasts calculated for the proportion of silage in the TMR were not significant ( P >0.05). (Maekawa et al., 2002)

PAGE 45

33 Table 2-3. Effects of forage particle size and grain fermentability on chewing activity, ruminal pH, and ruminal VFA profile in midlactation cows. Particle size ---------Chewing--------Ruminal VFA --Ruminal pH-Dieta Mean mm Eatb Rumc TCTd m M A:Pe Mean h<5.8f HMCFS 3.0 10.1 63 24.9 161.5 1.60 5.72 14.3 HMCCS 6.0 9.8 96 30.6 148.4 1.90 5.98 7.2 DCFS 2.8 9.8 54 21.7 151.1 2.23 5.90 9.3 DCCS 6.3 12.2 83 30.9 144.9 2.45 6.07 5.5 SED 0.8 6 1.7 5.6 0.12 0.08 1.4 Forageg 0.09 0.00010.00010.030.03 0.0006 0.0001 RFCg 0.07 0.03 0.24 0.100.0001 0.02 0.003 Forage x RFCg 0.02 0.60 0.16 0.400.68 0.39 0.11 aHMCFS = High-moisture corn and fine silage, HMCCS = high-moisture corn and coarse silage, DCFS = dry corn and fine silage, DCCS = dry corn and coarse silage, SED = standard error of the difference. bMinutes per kilogram dry matter intake cMinutes of rumination per kilogram NDF intake dTotal chewing time, minutes per kilogram dry matter intake eRuminal acetate to propionate ratio fHours per day below pH gP -value for effect of variable ( Krause et al., 2002; Krause and Combs, 2003) Table 2-4. Nutrient intake a nd digestion coefficients from giraffe fed all-hay diets. ----------------------Intake---------------------Digestion coefficient Treatment OM OM CW CP OM CW CP Units % BW kg/kg BW0.75 kg/kg BW0.75 g/kg BW0.75 % % % Grass 0.45 0.025 0.019 1.20 57.11 57.51 86.01 Alfalfa 0.89 0.049 0.026 10.90 60.70 55.92 93.76 Alfalfa SE 0.01 0.006 0.006 1.40 6.19 2.19 1.50 Values for 1 and 3 giraffe on ad libitum grass and alfalfa hay diets, respectively. OM= Organic matter; CW = cell wall; CP = crude protein Modified from Foose (1982).

PAGE 46

34 Table 2-5. Chemical composition (DM basis) of five browse plants grown at Busch Gardens in Tampa, Florida. Item Hibiscusa Bambooa Acaciaa Acaciab MulberryaFalse Acaciaa False Acaciab (DM) 18.7 47.7 39.7 46.8 27 28 52.2 Ash 16.87 11.6 14.23 9.58 11.48 7.88 6.05 NDF 53.9 65.1 33.7 46.1 33.6 36.4 56.7 NFC 15.8 12 37.2 34.8 34.6 43.3 28.9 Sugar 3.2 6.9 6.5 4.6 13.2 7.1 4.4 Starch 3.9 0.3 0.5 0.8 3.1 0.7 2.1 CP 15.4 12.9 16.5 13.6 25.7 18.3 10.2 NDICP 8.4 5.6 5.2 7.2 9.4 9.1 4.3 Ca 3.31 1.08 4.23 2.99 2.16 1.9 1.63 P 0.77 0.28 0.24 0.24 0.4 0.22 0.17 NDF = Neutral detergent fiber; NFC = non-fiber carbohydrates; NDICP = neutral detergent insoluble crude protein. aLeaf bLeaf + stem Samples collected in October, 2004. Unpublished data courtesy of Busch Gardens, Tampa.

PAGE 47

35Table 2-6. Influence of supplemental carbohydrate source fed in co mbination with 0.122% BW/ day of degradable intake protein on ruminal fermentation characteristics. ------------Orthogonal contrasts, P -value------------Control Starch MonoGlucose vs vs vs vs Component Control Starch Glucose Fructose Sucr ose SEM supplement sugar disaccharide fuctose pH 6.56 6.13 6.16 6.29 6.22 .04 <.01 .04 .97 .03 NH3 N, m M .31 2.42 2.99 2.85 1.88 .47 <.01 .79 .10 .83 OA, m Ma 70.8 98.9 96.0 89.0 89.1 2.91 <.01 .05 .37 .12 Acetate 73.5 69.5 61.5 59.8 59.7 .71 <.01 <.01 .34 .11 Propionate 14.0 16.4 14.1 14.2 14.4 .31 .05 <.01 .40 .81 Butyrate 10.6 10.3 17.5 18.9 19.5 .61 <.01 <.01 .11 .13 Isobutyrate .63 .82 .72 .66 .64 .05 .14 .02 .42 .39 Valerate .46 1.43 1.74 1.73 1.72 .06 <.01 <.01 .30 .92 Isovalerate .60 1.13 .99 .94 .90 .08 <.01 .07 .54 .71 Lactate .31 .45 3.52 3.82 2.95 .58 <.01 <.01 .33 .72 aOA = Total organic acids (VFA + lactate) VFA concentrations presented as mol/100 mol (Heldt et al., 1999)

PAGE 48

36 CHAPTER 3 EFFECTS OF ALTERING THE PHYS ICAL FORM AND CARBOHYDRATE PROFILE OF THE DIET ON CAPTIVE GIRAFFE Introduction Physical characteristics of feeds, particul arly of the fiber fraction, can influence the efficiency of use of a ruminant’s diet. Pr oper ruminal function in domestic ruminants is maintained by consumption of sufficient “physi cally effective fiber” (peNDF), which is determined by the neutral detergent fiber (NDF ) concentration and the ability of the feed to promote a chewing response in the animal (Mertens, 1997). The appropriate physical form, digestibility, and quantity of peNDF for concentrate selectors (CS) such as giraffe have not been determined. It has been propos ed that peNDF for CS includes particles of polygonal shape similar to browse, rather than the needle-like fibers derived from grasses (Clauss et al., 2002). Captive giraffe fed a tr aditional hay/ concentrate diet may face a nutritional dilemma: consume a high proportion of hay, increasing ruminal fill and decreasing intake, or consume a high proportion of concentrates, increasing potential risk of ruminal acidosis (Clauss et al., 2002). Nonfiber carbohydrates (NFCs) can be di vided into sugars (mono-, di-, and oligosaccharides), starch, organic acids, a nd neutral-detergent soluble fiber (NDSF), which includes carbohydrates such as pectin (Hall et al., 1999). The NFC fractions differ in fermentation characteristics and end products. Sugar (sucrose), starch and pectin have been shown to differ in vitro in both maximu m microbial protein yiel ds and in temporal pattern of that yield (Hall and Herejk, 2001). When fermen ted in the absence of other

PAGE 49

37 NFC sources, sugars have produced higher butyrate concentrati ons than starch (Heldt et al., 1999; Strobel and Russell, 1986). Str obel and Russell (1986) noted that unlike fermentation of sugars or starch, pectin ferm entation did not result in detectable lactate production at either neutral (6.7) or low (6.0) ph media. Fermentation of pectin or feeds high in NDSF has increased acetate concentr ations both in vitro (Ariza et al., 2001; Strobel and Russell, 1986) and in vivo (Ben-Ghedalia et al., 1989). Although the NFC fractions of giraffe diet s in nature have not been quantified extensively, analysis of leaves collected fr om five tree and shrub species used for enrichment of zoo animal diets showed greater DM concentrations of sugars (6.5 – 13.2) than starch (0.3 – 3.1) in four of five species (R. Ball, personal communications). Acacia leaves contained 37.2% NFC, 6.5% sugars, and 0.5% starch, sugge sting that NDSF, organic acids, and analytical may account for approximately 81% of the NFC fraction (R. Ball, personal communications). Consump tion of feeds high in NDSF by wild giraffe could explain why when (Clemens et al., 1983) examined molar propo rtions of volatile fatty acids (VFA) in the digestive tracts of 16 free-ranging African ruminant species, the highest ruminal proportion of acetate and the second highest acetate: propionate ratio were found in giraffe. Collectively, thes e observations raise questions about the importance of dietary NFC complement in maintaining optimal ruminal fermentation, and the response of captive giraffe to the various NFC fractions. Nutrient supply to the ruminant animal is not dependent upon diet composition alone. The amounts and proportions of offered dietary items th at the animal chooses to consume, digestion characteris tics and interactions of consumed dietary components, and nutrient absorption also must be considered. It has been suggested that the wasting and

PAGE 50

38 sudden death widely reported in captive gi raffe (Fox, 1938; Chaffe, 1968; Fowler, 1978; Strandberg et al., 1984; Flach et al., 1997; Junge and Bradley, 1993; Ball et al., 2002) may be a result of a functional energy defici ency ( Fowler, 1978; Ball et al., 2002). The reports of ruminal acidosis (Clauss, 1998; Cl auss et al., 2002b), fermentative gastritis or rumenitis (Fox, 1938; Ball et al., 2002), gast rointestinal ulceration (Fox, 1938; Fowler, 1978) and pancreatic pathologies ( Fox, 1938; Fowler, 1978; Lechowski et al., 1991; Ball et al., 2002) also may be indicative of cons umption of an unbalanced diet or altered fermentation of dietary components in captive giraffe. Compounding this problem, decreased ruminal absorptive surface area that has been reported in captive giraffe (Hofmann and Matern, 1988) may impair nutrient absorption. The objective of this study wa s to examine the effects of modified dietary physical form and NFC profile compared to a commerci al diet on captive giraffe intake, nutrient digestibility, behavior, and blood paramete rs. The hypothesis was that modifying supplement physical form and NFC profile w ould alter the intake nutrient digestion, behavior, and blood parameters of captiv e giraffe fed a hay + supplement diet. Materials and Methods Design This study was conducted from August 2002 through February 2003 at Busch Gardens in Tampa, FL (BGT). The desi gn was a modified reve rsal study using two treatments and six animals in seven 21-day periods (Table 3-1). The study was conducted according to protocols approved by the Institutional Animal Care and Use Committees of the University of Florid a and Busch Entertainment Corporation.

PAGE 51

39 Giraffe Six non-lactating adult female reticulated giraffe ( Giraffa camelopardalis reticulata ) were used. Body weight (BW) ra nged from 576 to 715 kg (average 637 + 42.6 kg). Giraffe were assigned identifica tion numbers G1 through G6, in order of entrance to the study. Three giraffe (G2, G4 and G6) were wild-caught, and estimated to be approximately 22 years of age at the tim e of the study. Three giraffe were born at BGT and aged 12 (G1), 7 (G3), and 4 (G5) years at time of entrance to the study. Pregnancy status of the animals varied. Two giraffe were in the third (G1) and second (G3) trimesters of pregnancy duri ng their time on study, and the remaining four giraffe were non-pregna nt. Nine and a half months pr ior to entrance to the study, G5 delivered a large (74 kg), rapidly growing cal f. The calf was weaned through progressive separation two to five days prior to G5Â’s entrance to the study. Facilities Study giraffe were individually housed in two adjacent pens, which allowed continual access to visual and tactile contac t with conspecifics. Pen dimensions were approximately 22 and 21 meters (east to west) by 8 and 9 meters (north to south) in pens 1 and 2 respectively. Flooring in each pen consisted of a roughed concrete pad of approximately 9 by 9 meters spanning the west end, and sand in the remainder of the pen. A shade shelter sloped east to west (9 and 7 meters high respectively) and covered the western 1/3 of each pen. Radiant heaters at tached to the bottom of the shade shelter provided heat when temperatures fell below 10 C. Pen 1 was equipped with a second shelter adjacent to the west end of the pe n that provided feed troughs with additional protection from driving rain. Feed in pen 1 suffered less weather damage than feed in pen 2 for one collection day in period 1 as a result. Water was provided in a 189-litre tub

PAGE 52

40 at ground height in pen 1, and an automa tic waterer (Nelson Manufacturing Company, Cedar Rapids, IA) at approximate ly 1.8 meters high in pen 2. Hayracks with catch pans and supplement feed tubs were mounted to the center of the west wall in each pen. Feeders were only accessible to the an imal in that pen. Hayracks and supplement feeders were placed adjacent to one another, with the top of each mounted to the top of the wall at appr oximately 3.2 meters high. At the onset of period 3, the supplement feed pan in pen 2 was lowered 15 cm to accommodate the shorter height of G3, and remained at th is height for the remainder of the study. Diets The standard BGT giraffe ration consisted of ad libitum alfalfa hay, fresh browse as available, and two concentrates, Mazuri Browser Breeder and Purina Omelene 200 (Purina Mills, St. Louis, MO). Concentrates were fed together in a 75:25 ratio (as-fed basis) at approximately 1% of gira ffe body weight as per manufacturerÂ’s recommendations for Mazuri Browser Breeder. This 75:25 grain mix (GF) was used as the control diet for comparative evaluation to the coarse, non-pelleted experimental browser feed (EF). Ingredient composition (DM basis) of EF was 35% sugar beet pulp, 18% cottonseed hulls, 13% molasses, 11% soybe an meal, 10% alfalfa meal, 6% mineral mix*, 4% heat-treated soybean meal (Soy Pl us) and 3% sucrose (table sugar). *Mineral mix was 94.5% DM, and was composed of the following ingredients (percentage inclusion on air dry basis): dicalcium phosphate (39.5%), calcium carbonate (8.76%), urea (8.76%), salt (5.25%), sodium bicarbonate (17.5%), potassium carbonate (9.65%), selenium 1% (0.12%), cobalt sulfat e (0.02%), copper sulf ate (0.08%), zinc sulfate (0.08%), iron sulfat e (0.30%), manganese sulfat e (4.38%), magnesium oxide (4.38%), ethylenediamine dihydroi odide (0.003%), vitamin A 650 x 106 IU/ kg (0.03%), vitamin D3 400 x 106 IU/ kg (0.01), and vitamin E 500 x 103 IU/ kg (0.80%). Mix contained: 84% ash, 24% CP as non-protei n N, 9.9% Ca, 8.3% P, 6.9% Na, 3.2% Cl, 2.6% Mg, 4.7% K, 0.96% S, 68 ppm Co, 203 ppm Cu, 20 ppm I, 5270 ppm Fe, 14118 ppm Mn, 12 ppm Se, 2060 ppm Zn, 88452 IU/# Vit. A, 18144 IU/# Vit. D3, 1814 IU/# Vit. E.

PAGE 53

41 The EF was formulated to be equivalent to GF in minerals and vitamins A, D, and E, and differed from GF and other commonly-fed br owser feeds by containing less starch, more sugars and soluble fiber, and small, heavily lignified particles (cottonseed hulls) to modify the fiber size and texture of the diet The experimental supplement was mixed in 178 kg batches by study personnel at the Univer sity of Florida Dairy Research Unit (Hague, FL). Each animal was housed individually and fed EF or GF ad libitum for 21 days, and then received the other feed supplement in the subsequent 21-day period, so that each animal received each diet. Alfalfa hay, water, and salt were offered ad libitum throughout the study. Sample Collection and Analyses Feedstuffs and intake Amounts of all feedstuffs offered and refu sed were weighed and recorded daily. Subsamples of all feedstuffs were collected. In periods 1 through 3, two to three flakes of alfalfa hay from a bale fed (and considered representative of alfalf a offered during that period) were retained for an alysis. In periods 4 through 7, a bale corer was used to sample five to seven of the bales fed dur ing that period. On days 15-21 of each study period, th total amount of hay and supplement orts for each giraffe were collected and frozen at 23 C until analysis. Dry matter (DM) content of supplement fed and daily supplement orts were obtained by drying duplicate subsamples of two to three grams each at 105 C in a forced air oven for approximately 36 hours, until a cons tant weight was achieved. Results were

PAGE 54

42 used to calculate daily dry matter intake (D MI) for each animal. Subsamples of offered supplements and supplement orts were dried at 55 C in a forced-air oven until constant weight was achieved. Subsamples were reta ined for particle size analysis, and the remaining dried sample was ground through a Wiley mill (A.H. Thomas, Philadelphia, PA) to pass through a 1-mm screen. In order to obtain representative subsam ples for DM determination, hay offered in periods 1 through 3 and total daily hay orts for each animal were chopped with a paper cutter to roughly 2.5 cm, th en blended and subsampled following the procedure of Van Soest and Robertson (1985). One subsampl e of approximately 10 grams was used to determine DM. Remaining sample was dried in a forced-air oven at 55 C, ground through a Hammer mill (Smalley Manufacturin g Co., Manitowoc, WI) using a 0.635-cm screen, mixed and subsampled. Subsamples were ground through a Wiley mill using a 1mm screen. Ground samples were analyzed for DM at 105 C and organic matter (OM) by ashing overnight (>8 hours) at 512 C in a muffle furnace (Sybron Corporation, Dubuque, IA). Daily orts were composited on a DM basis by giraffe by period. Batches of EF fed during the collection week of each period were composited by period. Mazuri Browser Breeder and Purina Omelene 200 fed in each period were analyzed separately for nutrient content, and results used to cal culate the nutrient co mposition of GF (75% Mazuri Browser Breeder, 25% Purina Omelene 200) fed in each period. Feed offered and composited orts were anal yzed for DM, OM and nutrient content. Neutral detergent fiber (NDF) and NDF orga nic matter (NDFOM) were determined using heat-stable -amylase (Goering and Van Soest, 1970; Van Soest, 1991). Acid detergent fiber (ADF) (Goering and Van Soest, 1970), s ugar (Hall, 2001), starch (Hall, 2001),

PAGE 55

43 lignin (Goering and Van Soest, 1970), and mineral (using Perkin Elmer 3300 XL ICP manufactured by Perkin Elmer, Shelton, CT ) (Association of Official Analytical Chemists, 1990) content were analyzed by Cumberland Valley Analytical Services, Maugansville, MD. Neutral detergent sol uble fiber was determined according to the method of (Hall et al., 1999). Crude pr otein (CP) as N x 6.25 was determined by combustion analysis (Association of Offici al Analytical Chemists, 1990) using Macro Elementar Analyzer (vario MAX CN, Elementar Analysensysteme GmbH, Hanau, Germany). Nutrient intake was calculated as: kg nutrient offered – kg nutrient refused. Particle size of supplements offered was analyzed using U.S.A. Standard Testing Sieves (Fisher Scientific Company, Pittsburg, PA) 5, 10, 18, 23, 60, and 120 (pore sizes of 4, 2, 1, 0.5, 0.25, and 0.125 mm). Samples of approximately 10 g DM were soaked in 300 ml of deionized water for 2 hours. Hydrat ed samples were quantitatively transferred to stacked sieves. The top sieve was rinsed for 5 minutes using a three-jet water sprayer calibrated to a flow of 1 L per minute (D Mertens, personal communication), then inverted and contents rinsed into a Gooch crucible under vacuum. This procedure was repeated for each sieve. Crucibles were dried at 105 C for approximately 36 hours to determine DM retained on each screen. Results were used to calculate modulus of finess (MOF) according to Poppi et al. (1980). Fecal collection and analysis Attempts at total fecal collection bega n on day 16 in period 3, and day 15 of each remaining period, and lasted for 56 108 c onsecutive hours. Dura tion of total fecal collection in a given period was often limited by adverse weather conditions or animal behavior that resulted in sample loss or safety risk to the an imal or staff. In period 3,

PAGE 56

44 attempts to collect the total daily fecal pr oduction of G3 were considered unsuccessful due to excessive pacing by that animal. To tal collection was not attempted on G3 in period 4. Samples were collect ed in gallon Ziploc bags a pproximately six times per day, as staff became available, and for two hour s overnight on days 16 and 17 in period 3, and days 15 and 16 in remaining periods Samples were frozen at -23 C for subsequent analysis. Fecal samples from each giraffe were composited by each 24 hours of collection, yielding two consecutive 24-hour composites fo r each giraffe in each study period, with the exception of G3 in P4, when total collection was not attempted. Compositing procedure was as follows. Samples were thawed in their original ba gs overnight at room temperature. Each individual bag of fecal material was weighed on a Mettler PE 3600 top-loading balance. Bag contents were emp tied into a plastic cont ainer. All non-fecal organic debris and as much contaminating sand as possible were removed from the sample and returned to the or iginal bag. Bag and debris weight was subtracted from initial weight to calculate sample wet weight. Sample was mixed manually, and subsamples of 10% + 0.02% of original wet weight were taken from each bag for each composite. Two composites were retained fo r each giraffe by day, and frozen at -13 C until analysis. Composite 1 was dried for analysis of DM OM, NDF and CP. The thawed sample bag was laid flat and rolled with a plasti c cylinder to crush f ecal pellets and blend material, facilitating accuracy of subsampli ng. Duplicate subsamples of approximately 23 g each were used to determine DM and OM and the remaining sample was dried in a forced air-oven at 55 C and ground to pass through the 1-mm screen of a Wiley mill.

PAGE 57

45 Crude protein as N x 6.25 was determined in duplicate on 0.25 – 0.35 g samples by combustion analysis (Association of Offici al Analytical Chemists, 1990) (Vario MAX CN, Elementar Analysensysteme GmbH, Hana u, Germany). Duplicate samples of 0.7 g each were used to determine NDFOM using heat-stable -amylase (Goering and Van Soest, 1970; Van Soest, 1991 ). Percent dige stibility of NDFOM and apparent percent digestibility of OM and CP were calculate d as: (mean g nutrient consumed per day) – (mean g nutrient defecated per day) / (mean g nutrient consum ed per day), using intake values from days 14 through 20 in each period and fecal samples collected for 48 consecutive hours on days 16 through 18 in periods 3 and 6, and days 15 through 17 in remaining periods. Nutrient kg digested wa s calculated as: nutrient kg consumed x percent digestibility. Composite 2 was crushed manually, subsam pled for DM and OM, and analyzed wet for CP, to examine for volatile nitr ogen (N) loss due to drying and grinding. Composite 1 and composite 2 samples were analyzed for CP using macro-Kjeldahl analysis (Association of Offici al Analytical Chemists, 1990). Behavior On days 13 through 15 of each period, beha vior was recorded for 48 consecutive hours by on-site trained observe rs. Rumination, consumption (supplement, hay, water, salt and sand), oral stereo types (tongue play, licking metal and licking non-metal inanimate objects), tactile contact between gi raffe, and locomotor behaviors (standing, walking, lying down with head erect and lying down with head on flank) were recorded every sixty seconds using instantaneous sampling (Martin and Bateson, 1986). Results were analyzed for differences in the numb er of minutes over 48 hours engaged in each individual behavior, total or al stereotypes, supplement + hay consumption, supplement +

PAGE 58

46 hay consumption + rumination, and total oral be havior (all oral ster eotypes + total feed, water, salt and sand consumption + rumination). Body weight and blood samples A 2.4 meter high solid wooden chute syst em adjacent to pen 1 provided manual restraint for blood collection. The chute was equipped with a scale for measuring body weight. At approximately 0800 hours on day 1 of the first period an animal entered the study and day 21 of each period, weight and bod y condition scores (scored by C. Kearney on a scale of 1 through 8) were recorded a nd blood samples collected on each animal before feeding. Blood samples were collect ed via jugular venipuncture using manual restraint (chute system). Blood samples were collected on all giraffe using clot tubes and three anti-coagulants: ethylenediaminetetraac etic acid (EDTA), sodium heparin (NaH), and sodium citrate (NaC) (Vacutainer Brand tubes by Becton Dickinson, Franklin Lakes, NJ). Blood was collected using an 18or 14-gauge needle, 30-inch extension set and 35cc syringe. Blood samples were transferred into the tubes, pl aced on ice, and returned to the zoo hospital laboratory w ithin two hours of collection. Sodium citrate samples were placed on a rocker until shippi ng for fibrinogen analysis. Complete blood count (on whole blood pr epared with EDTA) (Automated Cell Counter) and serum chemistry profiles (Hita chi 747-200 Chemistry Analyzer, Hitachi, Hitachi-NAKA, Japan) were analyzed by An tech Diagnostics (Largo, FL). Serum chemistry profiles evaluated glucose, urea nitrogen, creatinine, to tal protein, albumin, total bilirubin, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, cholesterol, Ca, P, Na K, Cl, Mg, globulin, lipase, amylase, triglycerides, creatinin e phosphokinase, gamma-glutamyltransferase, lactate dehydrogenase, and calculated osmolality. Additional plasma and serum were pipetted

PAGE 59

47 into scintillation vials and frozen at –80 C for later analysis. Plasma with NaH was analyzed for insulin (double antibody radioimmunoassay proce dure; Soeldner and Sloane, 1965), and non-esterified fatty acids (NEFA) (NEFA-C kit; Wako Fine Chemical Industries USA, Inc., Dallas, TX; as modi fied by Johnson and Pete rs (1993) at the University of Florida. Statistical analysis Data were analyzed by the MIXED pro cedure of SAS (1999). Animal response data were analyzed using a st atistical model that included animal, period, and diet with animal as a random variable. Supplement nutri ent content and fecal nutrient content were analyzed using the model “nutrient = diet”. Results are reported as least squares means with standard errors as determine with PROC MIXED. Pearson correlation coefficients were calculated using the PROC CORR procedur e of SAS. Due to the small number of animals in this study (n=6), significance was declared at P <0.10, and tendency at 0.10< P <0.15. Results and Discussion As planned, the EF diet contained greater concentrations of sugar ( P <0.001) and NDSF ( P <0.001), and decreased con centrations of starch ( P <0.001) (Table 3-2). Concentrations of ash, lignin, and NDFOM were also greater in EF, and CP concentrations tended to increase as well, a nd variable differences occurred in mineral concentrations (Table 3-2). S upplement particle size differed ( P <0.001) between diets. The EF supplement was courser than GF, w ith a greater MOF value (4.97) than GF (3.15).

PAGE 60

48 Intake Average daily DM intake varied greatly among the six animals for both alfalfa hay (0.12 to 3.94 kg/day) and supplement (2.87 to 9.26 kg/day) consumption, but did not differ between diets (Table 3-3). As giraffe shifted from consuming supplement GF to EF, starch intake decreased ( P =0.052) from 0.93 to 0.12 kg/ day, sugar intake tended ( P =0.115) to increase from 1.12 to 1.53 kg/ day, and NDSF intake increased ( P =0.074) from 0.85 to 1.19 kg/day. Consumption of ADF (1.83 vs. 2.23 kg/day) ( P =0.039) and lignin (0.33 vs. 0.50 kg/day) ( P =0.064) differed between animals consuming supplement GF and those consuming EF, respectively. Although supplements were formulated to be equivalent in concentration of minerals, CP, and NDF, analysis of the supplem ents fed revealed slight differences in these analytes (Table 3-2). As a result, treatment-associated differences for mineral intake (Table 3-3) may reflect differences in supplement mineral concentrations. Differences in supplement nutrient concen tration and intake may account for the increased consumption of NDFOM and ash when giraffes were offered EF. There was a tendency for greater CP consumption by gira ffe offered EF than those offered GF ( P =0.136). Pellew (1983) estimated intake of wild giraffe in the Serengeti by recording the number of bites during feeding bouts, then multiplying by mean bite mass. Mean bite mass was first estimated by hand clipping foliage to simulate giraffe browsing behavior, then corrected using observations of th e number of bites taken by captive giraffe consuming a known quantity of browse during timed feeding periods. Daily DM intake was estimated as 19.0 kg for males and 16.6 kg for females. Using average live weights of 1200 kg and 800 kg, as reported by Dagg and Foster (1976), mean daily DM intake

PAGE 61

49 was estimated at 1.6 and 2.1% of BW for male s and females, respectively. Dry matter intake as a percentage of BW in th e present study ranged from 0.69 to 1.66% and averaged 1.22% (+ 0.28), much lower than intake re ported for wild giraffe (Pellew, 1983), but similar to DM intake by non-lactat ing captive giraffe reported by Baer et al. (1985) of (1.22% of BW) and by Clauss et al. (2001) of (0.97 to 1.28% of BW). Digestibility Several challenges occurred w ith attempted total fecal colle ction. In period 1, total collection was attempted on days 15 and 16, but was considered suspect on day 16 because of possible sample loss due to hea vy rain. When laboratory analyses yielded unusually high digestibility values for day 16, it was concluded that total collection likely had not been achieved, and this day was exclude d from data analysis. Digestibility data from G3 in period 3 and G4 in period 4 we re also excluded from data analysis, since collection records, high sample ash content, and high digestibility re sults suggested that excessive trampling had prevented total collec tion of fecal material and given excessive contamination with sand. Since collection was not attempted on G3 in period 4, no digestibility data is reported for period 4. Because samples were collected from sa nd-bedded pens, sand contamination was a continual complicating factor. Although samples were “clean ed” manually at the time of collection and again prior to compositing, ash content of f ecal composites ranged from 20 to 63% of sample DM (20 to 46% in sample s included in results). Sand contamination increased subsampling error, which often in creased variation between duplicates when samples were analyzed for OM, CP, and especial ly NDFOM at the University of Florida. Subsampling of fecal samples and analysis of NDFOM was evaluated at the U.S. Dairy Forage Research Center (US DA-ARS, Madison, WI), where the same difficulties were

PAGE 62

50 encountered. In the end, any samples not meeting the Horwitz criteria (Horwitz, 1982) for difference between duplicates were re-a nalyzed in duplicate at the University of Florida, and the mean of f our values was reported. Previous attempts at determining appare nt digestibility of DM, NDF, and CP in captive giraffe fed hay/concentrate diets usi ng acid detergent lignin, acid insoluble ash, indigestible NDF and alkanes as markers have been summarized by Clauss et al. (2001). Mean apparent digestibility of CP, OM, a nd NDFOM in the present study fell within the range of apparent digestibility of CP (59.7 to 82.4%), DM (52.0 to 85.2%), and NDF (32.5 to 74.8%) previously reported for captive gi raffe (Clauss et al., 2001). Digestibility of NDFOM averaged 55.5 + 4.0%, and was not affected by treatment (Table 3-3). The kg of NDFOM digested increased ( P =0.077) by 0.25 kg on EF, and was likely due in part to the 0.47 kg increase in NDFOM intake on EF. Apparent OM digestibility averaged 72.0 + 4.5%, and decreased numerically ( P =0.206) on the EF diet. Apparent CP digestibility was numerically 10 percenta ge units greater for GF than EF ( P =0.151), but apparent kg of CP digested differed by onl y 0.06 kg (Table 3-3). The EF supplement included 4% heat-treated soybean meal, contai ning protein which is less degradable in the rumen. We do not know if a ruminally unde gradable protein source was present in GF. This potential difference between supplem ents may have contributed differences in true CP digestibility between diets. Altern atively, an increase in yield of ruminal or hindgut microbes may have facilitated an increa sed presence of microbial CP in the fecal material of giraffe consuming EF. The CP concentrations of feces fr om giraffe fed EF were greater ( P <0.004) than concentrations for giraffe fed GF using all th ree analytical technique s (Table 3-4). When

PAGE 63

51 fecal samples were analyzed for CP in dry and wet forms, CP concentrations were generally higher in the wet samples by approxi mately two percentage units, suggesting a loss of volatile nitrogen during the drying proces s. Apparent CP digestibility calculated using dried feces may have been overestimated as a result. Behavior The effect of dietary treatment on giraffe behavior is reported in Table 3-5. The number of minutes engaged in ruminati on and hay consumption over 48 hours was not affected by treatment. Time engaged in supplement consumption was 2.3 times longer by giraffe offered EF than by giraffe offered GF ( P =0.063), and total feeding time tended ( P =0.100) to be greater by giraffe offered EF. The tendency ( P =0.124) for increased time spent in eating and rumination by giraffe fed the EF diet has potential to have increased saliva flow and ruminal buffering. Leuthold and Leuthold (1972) reported on the diurnal time budgets of wild giraffe. Female giraffe spent 53.1% and 15% of daylight hours engaged in feeding and rumination, respectively (Table 3-6). By c ontrast, giraffe in this study spent 17.7% and 22% of time feeding and ruminating It is interesting to note that oral stereotypes, which have not been observed in wild giraffe, in creased the total time captive animals spent engaged in oral behavior from 39.7% for feeding and rumination alone to 54.1%, which bears more similarity to the 68.1% of time wild giraffe spent engaged in feeding and rumination. Oral stereotypy appears to be the most pr evalent stereotypic be havior observed in captive giraffe. A survey of giraffe and / or okapi-holding American Zoo and Aquariums Association (AZA) accredited in stitutions by Bashaw et al. (2001) yielded data on the occurrence of stereotypes in 214 giraffe a nd 29 okapi in 49 institutions. The most

PAGE 64

52 prevalent stereotypic behaviors were repetitive li cking of non-food objects (referred to as “licking”) and pacing. Licki ng was reported in 72.4% of gi raffe + okapi, and pacing was reported in 29.9%. Additional behaviors repo rted in 3.2% of animals included head tossing, self-injury, and tongue playing. An a ttempt to reduce incide nce of stereotypic behavior had been made by 51.7% of respondi ng institutions, with reported success in 51.9% of these institutions. Tarou et al (2003) attempted using Bitter Apple, a chemical spray used to deter undesirable chewin g in horses, to deter stereotypic licking in three captive giraffe. Animals simply shifte d licking behavior to a non-treated area. Bashaw et al. (2001) suggested that lick ing in captive giraffe and okapi may be related to feeding motivation. In the presen t study, oral stereotype s were recorded as three separate oral behaviors: repetitive licking of metal objects, repetitive licking of nonmetal objects, and tongue play unassociate d with feeding, rumination, drinking, or licking. Licking metal was the most prevalen t oral stereotype (mean = 258 minutes / 48 hours), followed by tongue play (mean = 105 minutes / 48 hours) and licking non-metal objects (mean = 20 minutes / 48 hours). The amount of time spent engaged in tongue play tended ( P =0.119) to be twice as long in giraffe offered GF (Table 3-5). Although all six animals exhibited each of the individua l stereotypes, they varied in individual preference for metal licking or tongue play (F igure 3-1). The number of minutes over 48 hours spent engaged in total oral stereotype behavior ranged from 209 to 661, with a mean of 383 + 127. Despite the increase in time e ngaged in feeding behavior, minutes engaged in total oral stereoty es decreased only numerically ( P =0.223) from 433 to 318 minutes in giraffe offered EF.

PAGE 65

53 During this study, giraffe were repeatedly observed drooling or swallowing while engaged in oral stereotype behavior. Although this observa tion is purely subjective in nature, it suggests that oral stereotypy may also facilita te some amount of ruminal buffering. Blood Measures Few blood parameters were affected by diet (Table 3-7). Animals consuming GF had higher ( P =0.028) blood glucose values (99.0 mg/d l), compared to those consuming EF that showed values (82.3 mg/dl) more si milar to the range reported for domesticated ruminants (40-80 mg/dl) (Swenson, 1984). Sw itching captive giraffe from GF to EF decreased starch consumpti on and increased NDSF consump tion. Ruminal fermentation of NDSF rather than starch has increased ace tate and decreased propi onate concentrations in ruminal fluid in domestic livestock studies. Since acetate may be utilized directly as an energy source or used as a lipogenic substrat e, whereas propionate is converted largely to glucose, a potential NFC profile-associated shift in ruminal VFA production offers one possible explanation for the decreased blood gl ucose concentrations in giraffe fed EF. Despite the similarity in apparent kg of CP digested, blood urea nitrogen showed a numerical decrease ( P =0.166) from 20.6 mg/dl in giraffe fed GF to 16.6 mg/dl in those fed EF. Comparatively for animals consumi ng EF, this may reflect a shift in dietary protein utilization towards pr oliferation of microbial mass, decreased degradation of body protein, or decreased degradati on of absorbed amino acids. Non-esterified fatty acids (NEFA) decr eased numerically for animals consuming EF ( P =0.282), and decreased on EF in all giraffe except G1 (Figure 3-2), who elected to consume a 99% supplement diet when fed EF. A BW gain of > 15 kg occurred in 5 of 6 giraffe when EF was fed, and only in 1 of 6 giraffe when GF was fed, but treatment-

PAGE 66

54 associated changes in BW varied am ong animals and were not significant ( P = 0.327) (Table 3-3). These observations raise the question of the basis for possibly decreased adipose tissue mobilization with diet EF. C ould part of this effect be the result of increased NDSF intake and increased mass of NDFOM digested altering ruminal acetate:propionate ratio, providing the an imals with more lipogenic nutrients? Blood glucose was positively correlated with kg of OM and CP digested and consumed (Table 3-8). Although BUN showed a positive correlation with apparent kg of CP digested, a stronger positive correlation existed between BUN and starch intake. Glucose and BUN were positively correlated with one another, and of the blood proteins measured, CPK was the only protein positivel y correlated with both glucose and BUN (Table 3-9). Increased blood concentrations of CPK have been associated with increased muscle catabolism. Ancillary Study Observations / Individual Animal Effects The giraffe originally designated to serve as G4 became ill at the time of entrance to the study. As a result, the study was halted for three weeks, and this animal eventually entered the study as G6. During the break in the study, G3 continued to receive the EF diet, which she had received in the previ ous period. Over the 6 weeks on EF, her body weight increased by 35 kg, from 542 to 577 kg. While pregnancy status may have been a contributing factor, her wei ght gain during three weeks on GF was only 3 kg. This observation, coupled with increased body wei ght and condition in a lactating non-study giraffe fed EF for several months suggest th at greater changes in animal response may have been observed had study periods been of longer duration. Fu rthermore, ancillary observations suggest that feeding EF to two lactating giraffe not on study may have increased milk production. When these females were switched from the normal giraffe

PAGE 67

55 ration (GF) to EF, the fecal consistency of their calves was loose the following day, and gradually returned to normal over the subsequent days. By using six non-lactating adult female gi raffe, this study contained the largest and most similar giraffe population reported in a captive feeding study. Even so, each animal proved to be a unique individual, exhibiting qua lities or behaviors th at no doubt affected study results. Giraffe 1 elected to consum e a diet of 98.7% supplement and 1.3% hay when fed the EF diet. Giraffe 2, an arthri tic older female with a historically strong exhibition of maternal behavior, was on the study for three periods, and received the EF supplement in both periods 1 and 3. The low mean voluntary intake for G2 in period 1 (0.69% of BW), which was lower than other reported intake values (0.86 to 1.66% of BW), may have been related to factors othe r than dietary treatment. During the eleven acclimation days in period 1 during which as -fed intake values can be considered reasonably accurate (days wit hout heavy rain), her daily as -fed intake averaged 7.01 kg, similar to mean DM intake (6.92 kg) during the collection week of period 3. However, mean daily intake on days 16-18 of period 1 decreased by nearly half, to 3.58 kg (3.31 kg of DM). This may have been related to e nvironmental factors rath er than diet. Sand erosion caused by heavy rain had made it necessary for G2 to step up onto the concrete pad in order to access feeders. While atte mpting to walk onto the concrete pad on day 18, G2 was observed “tripping” over a small ir rigation pipe that had been exposed by sand erosion. After the pipe was remove d during the morning feeding on day 19, G2 would not step up onto the concrete, and cons umed no feed that day. Consequently, day 19 was removed from analysis of intake re sults. During morning feeding on day 20, a sand ramp was constructed, and feeding be havior resumed. However, G2 did not

PAGE 68

56 approach the feeders after the evening feeding on day 20, appearing instead to be preoccupied with the birth and overnight keep er observations of a giraffe calf in an adjacent pen. Thus, intake amounts on day 20 (3.15 kg of DM) were similar to intake on days 16 through 18. Giraffe 3 had not been housed in a small pen prior to this study, and had a history of nervous behavior around hu mans. She spent a great deal of time pacing, displaying nervous behavior which may have affected intake and blood values, and certainly prevented total fecal collection. Giraffes 3 and 4 frequently engaged in tongue play with the mouth closed, rather than open, making the behavior difficult to detect from a distance. Consequently, oral stereotype behavior may have been underestimated or disproportionately reported in these animals. Giraffe 5 was on the study for three pe riods, and received GF supplement in periods 5 and 7. Prior to study entrance, th is primiparous female had been nursing a rapidly growing calf, and had shown a d ecline in body weight and condition. Body weight increased steadily from 611 kg at st udy entrance to 661 kg at study exit, and body condition score increased from 3.5 to 4.5 on a scale of 1-8, sugge sting that G5 was recovering from the demands of lactation dur ing her time on the study. Giraffe 5 also elected to consume hay as a substantially greater proportion (39 to 41%) of total DM intake than other giraffe. In short, individual animal vari ation may have played a large role in the results of this study. Diet Selection Both dietary physical form and carbohydrate profile play a role in development or prevention of ruminal acidosis. Voluntary in take patterns of low amounts of forage and high amounts of concentrate by captive giraffe ap pear similar to thos e associated with

PAGE 69

57 ruminal acidosis in domestic livestock. Hay: supplement intake ratios for giraffe in the current study ranged from 1:99 to 41:59 and averaged 21:79 (DM basis). Previously reported hay: concentrate intake ratio of tw o giraffe fed in the same pen was 26:74 (Baer et al., 1985). Another study using four gi raffe supplemented with limited amounts of browse in addition to lucerne hay and concen trates found forage:concentrate intake ratios of 25:75 to 51:49 (Clauss et al., 2001). The forage: supplement inta ke ratios of the 21 measures taken from these three studies av eraged 27:73 (DM basis) Decreasing dietary ratios of forage: concentrate from 40:60 to 30:70 has been used to induce subclinical ruminal acidosis in dairy cattle (Krajcarsk i-Hunt et al., 2002). Although giraffe are believed to be adapted to diets characterized by a rapid ruminal fermentation and nutrient absorption, a key factor in this adaptation is greatly increased ruminal surface area. Papillary development plays an important role in stabi lizing pH by preventing acid accumulation via absorption of ruminally-produced organic acids (Dirksen et al., 1985). The captive giraffe examined by Hofmann and Matern (1988) had approximately 11% of the ruminal surface area of that found in wild conspecifics, suggesting they had partially lost their ability to deal with rapid producti on of VFA in the rumen. Even pH reductions that do not reach the threshold for ruminal acidosis can alter the ruminal environment, and notable changes in fermentation characte ristics occur when pH decreases to <6.2 (Russell and Wilson, 1996). If hay: concentrate intake ratios of the eleven giraffe in the three aforementioned studies, low (relative to wild giraffe) feeding and rumination time by giraffe in the present study, and papillae development of the two giraffe cited in Hofmann and Matern (1988) are representative of captive giraffe in general, ruminal pH

PAGE 70

58 reduction, and quite possibly some degree of ruminal acidosis, is likely occurring in a substantial percentage of th e captive giraffe population. The lack of documentation of acidosis in cap tive giraffe may relate to the degree of pH reduction and the extreme difficulty in diagnosing non-catastrophic degrees of acidosis without rumen fluid collection (Garre tt et al., 1999), which is unlikely to occur in captive giraffe. Symptoms of subacute (p H<5.5) rather than acu te (pH<5.0) acidosis are “insidious and considerably less overt,” and incl ude decreased or variable feed intake, decreased efficiency of milk production, poor body condition, unexpl ained diarrhea, and episodic laminitis (Nocek, 1997), symptoms wh ich have occurred in captive giraffe, but have generally been attributed to other causes. The variation in se lective consumption of forage and concentrate noted in the pres ent study may dictate which animals are at greater risk of ruminal disord ers. Provision of mixed diets that reduce the ability of the animals to selectively consume c oncentrates may reduce this risk. Conclusions Captive giraffe offered EF as compared to those offered GF increased the amount of time engaged in feeding behavior to a leve l closer to that obser ved for wild giraffe, which also may have increased ruminal bu ffering via saliva production at the time of rapidly-fermenting NFC consumption. Decreased starch and increased NDSF contents of EF facilitated consumption of a carbohydrate prof ile that may bear a greater similarity to natural giraffe feedstuffs than typical zoo con centrate diets. A possible resultant shift in ruminal VFA production toward the high acetate : low propionate profile observed in wild giraffe could contribute to the obs erved decrease in blood glucose. The fact that few significant treatment eff ects were observed in this study may be a result of low animal numbers, individual an imal variation, and a relatively short time on

PAGE 71

59 treatment. The experimental supplement wa s formulated in an attempt to improve ruminal function and overall animal respons e, and while results suggest that this formulation may have some promise, furt her research is need ed for confirmation. Table 3-1. Design of study. Pen 1 Pen 2 Period Dates Giraffe Diet Giraffe Diet 1 08/20 09/09/02 2 EF 1 GF 2 09/10 09/30/02 2 GF 1 EF 3 10/01 10/21/02 2 EF 3 EF 4 11/12 12/02/02 4 EF 3 GF 5 12/03 12/23/02 4 GF 5 GF 6 12/24 01/13/03 6 GF 5 EF 7 01/14 02/03/03 6 EF 5 GF EF = Experimental supplement; GF = Control supplement

PAGE 72

60 Table 3-2. Chemical composition of alfalfa hay and supplements (d ry matter basis) fed to captive giraffe, and difference between supplements. ----------------------Supplem ent ---------------Alfalfa hay GF EF SE P values Dry matter (%) 90.8 88.3 85.2 0.29 0.001 Organic matter (%) 91.0 92.0 89.1 0.07 <0.001 Ash (%) 9.04 8.02 10.9 0.07 <0.001 Crude Protein (%) 19.2 17.1 17.6 0.25 0.138 NDFOM (%) 41.2 35.1 40.3 0.60 0.003 ADF (%) 33.8 23.0 26.0 0.41 0.002 Lignin (%) 7.12 3.32 5.67 0.12 <0.001 NDSF (%) 15.2 9.64 14.9 0.25 0.001 Sugar (%) 11.6 14.7 21.4 0.51 <0.001 Starch (%) 1.20 14.2 1.47 0.06 <0.001 Ca (%) 1.65 1.17 1.28 0.02 0.021 P (%) 0.22 0.79 0.65 0.02 0.007 Mg (%) 0.34 0.44 0.56 0.04 0.080 K (%) 1.63 1.21 1.91 0.03 <0.001 Na (%) 0.16 0.30 0.46 0.02 0.003 Fe (ppm) 97.8 413 674 98.9 0.003 Mn (ppm) 31.9 141 1097 75.4 <0.001 Zn (ppm) 37.6 160 282 31.0 0.050 Cu (ppm) 4.94 23.8 29.3 4.26 0.410 NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber EF (experimental supplement) was calculated to contain 0.29% S, 0.38 ppm Co, 387 ppm Fe, 1.0 ppm I, 0.77 ppm Se, 6.29 KIU/lb Vitamin A, 1.50 KI U/lb Vitamin D,and 118.2 3 KIU/lb Vitamin E. GF (control supplement) was composed of 75% Mazuri Browser Breeder and 25% Purina Omolene 200. Mazuri Browser Breeder is reported to contain 0.21% S, 0.37 ppm Co, 370 ppm Fe, 1.0 ppm I, 0.69 ppm Se, 16,250 IU/kg Vitamin A, 3,000 IU/kg Vitamin D3 (added), 240 IU/kg Vitamin E, 5.3 ppm Vitamin K, 12 ppm thiamin hydrochloride, 10 ppm riboflavin, 47 ppm niacin, 44 ppm pantothenic acid, 1070 ppm choline chloride, 1.5 ppm folic acid, 9.9 ppm pyridoxine, 0.21 ppm biotin, and 46 mcg/kg Vitamin B12. Purina Omolene 200 is reported to contain not less than 0.60 ppm Se and 3000 IU/lb Vitamin E. Table 3-3. Effects of dietar y treatment on mean daily dr y matter and nutrient intake, digestion of organic matter and crude protein (apparent) and NDFOM (true), body weight gain and body condition score. --------P-values-------Intake, DM basis GF EF SE Diet Period Total (% of BW) 1.20 1.25 0.10 0.294 0.123 Supplement (% of BW) 1.00 0.98 0.10 0.81 0.237 Hay (% of BW) 0.24 0.23 0.06 0.63 0.332 Total DM, kg/ day 7.60 7.91 0.66 0.527 0.273 Supplement, kg/ day 6.17 6.30 0.58 0.83 0.351

PAGE 73

61 Table 3-3. Continued. --------P-values-------Intake, DM basis GF EF SE Diet Period Hay, kg/ day 1.47 1.52 0.41 0.80 0.348 Organic matter, kg/ day 7.08 7.20 0.63 0.725 0.239 Ash, kg/ day 0.52 0.72 0.05 0.157 0.351 Crude protein, kg/ day 1.32 1.41 0.14 0.136 0.095 NDFOM, kg/ day 2.80 3.27 0.28 0.198 0.287 Acid detergent fiber, kg/ day 1.83 2.23 0.22 0.039 0.078 Lignin, kg/ day 0.33 0.50 0.05 0.064 0.275 NDSF, kg/ day 0.85 1.19 0.10 0.074 0.251 Sugar, kg/ day 1.12 1.53 0.11 0.115 0.272 Starch, kg/ day 0.93 0.12 0.04 0.052 0.675 Calcium, kg/ day 0.10 0.11 0.01 0.048 0.042 Phosphorous, kg/ day 0.05 0.04 0.00 0.185 0.329 Magnesium, kg/ day 0.03 0.04 0.00 0.176 0.201 Potassium, kg/ day 0.10 0.14 0.01 0.013 0.057 Sodium, kg/ day 0.02 0.03 0.00 0.112 0.219 Digestibility OM, % of DM 74.1 69.3 1.06 0.206 0.413 NDFOM, % of DM 56.0 54.6 1.02 0.513 0.331 CP, % of DM 75.5 65.7 2.22 0.151 0.593 OM, kg 5.09 4.91 0.42 0.143 0.063 NDFOM, kg 1.52 1.77 0.17 0.077 0.124 CP, kg 0.97 0.91 0.10 0.253 0.194 Body weight gain, kg 4.84 12.00 3.78 0.328 0.454 Body condition score 3.85 4.18 0.24 0.247 0.515 NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber; GF = control supplement; EF = experimental supplement Table 3-4. Effect of diet on fecal nut rient composition (dry matter basis). --------P-values-------GF EF SE Diet Period OM 62.8 64.9 6.94 0.3066 0.0181 NDFOM 42.2 44.4 6.47 0.1123 0.0011 CP (Dried feces)a 11.2 14.2 1.78 <0.0001 0.0017 CP (Wet feces) b 13.4 16.0 1.67 0.0003 0.0054 CP (Dried feces) b 11.6 14.5 1.44 0.0035 0.2830 GF = control supplement; EF = experimental supplement; NDFOM = neutral detergent fiber OM aCombustion analysis bKjeldahl analysis

PAGE 74

62 Table 3-5. Recorded behavior of captive giraffe consuming diffe rent supplements. ---------Supplement----------------P-values-------Behavior, minutes/ 48 hours GF EF SE Diet Period Standing 1567 1698 281 0.012 0.017 Walking 272 194 84 0.631 0.434 Lying 1009 906 87 did not converge Sleeping 19 13 3 0.374 0.475 Social contact 23 23 4 0.994 0.034 Rumination 609 677 51 0.307 0.322 Hay consumption 239 254 45 0.762 0.431 Supplement consumption 121 277 25 0.063 0.255 Total eating 359 530 42 0.100 0.312 Eating + rumination 953 1224 35 0.124 0.187 Drinking 58 54 1 0.416 0.589 Salt 1 0 1 0.416 0.589 Metal licking 255 212 70 0.390 0.357 Non-metal licking 17 22 7 0.716 0.493 Tongue play 161 80 39 0.119 0.302 Total oral stereotype 433 318 57 0.223 0.549 Wall foraging 37 36 14 0.983 0.816 Total oral behaviora 1481 1632 71 0.363 0.455 GF = control supplement; EF = experimental supplement aTotal eating + rumination + total oral stereotype + wall foraging Table 3-6. Percentage of time female gi raffe spent engaged in oral behaviors. ---------------------Captivea---------------------Wildb Minimum Maximum Mean SE Mean Feeding 9.1 38.2 17.7 2.16 53.1 Rumination 15.8 32.6 22 1.38 15 Oral steriotypes 7.3 23 13.3 1.18 NA Oral behavior 43 66.1 54.1 2.1 68.1 a6 giraffe over 48 hours. b5 giraffe during daylight hours. (Leuthold and Leuthold, 1978) Table 3-7. Effects of type of dietar y supplement on giraffe blood parameters. --------P-values-----Measure Units GF EF SE Diet Period Non-esterified fatty acids mEq/L 0.47 0.38 0.06 0.282 0.283 Insulin ng/ml 1.04 0.84 0.27 0.709 0.747 Glucose mg/dl 99 82.3 8.14 0.028 0.041 Blood urea nitrogen mg/dl 20.6 16.6 0.7 0.166 0.549

PAGE 75

63 Table 3-7. Continued -----P-values----Measure Units GF EF SE Diet Period Creatinine mg/dl 1.67 1.71 0.06 0.717 0.175 Total protein g/dl 8.92 9.03 0.32 0.76 0.731 Albumin g/dl 2.49 2.49 0.07 0.973 0.429 Total bilirubin mg/dl 0.12 0.16 0.01 0.295 0.356 Alkaline phosphatase U/L 158 148 17.6 0.516 0.386 Alanine aminotransferase U/L 11.3 7.51 1.51 0.244 0.506 Aspartate aminotransferase U/L 55.6 45.7 5.55 0.416 0.752 Cholesterol mg/dl 33.2 29.2 1.86 0.208 0.335 Ca mg/dl 8.55 8.95 0.29 0.309 0.577 P mg/dl 10.9 10.2 0.6 0.5 0.325 Na mEq/L 147 146 1.32 0.538 0.526 K mEq/L 4.58 4.86 0.13 0.183 0.327 Cl mEq/L 107 104 2.06 0.38 0.851 Mg mEq/L 2.17 2.07 0.5 did not converge Globulin g/dl 6.44 6.55 0.35 0.674 0.833 Lipase U/L 31.3 42.5 5.4 0.404 0.425 Amylase U/L 11.8 12.2 0.89 0.817 0.551 Triglycerides mg/dl 41.6 35.4 5.41 0.502 0.874 CPK U/L 248 258 69.8 0.942 0.737 GGTP U/L 18.5 19.7 3.23 0.799 0.511 Calculated osmolality mOsm/L296 291 3.01 0.39 0.523 Lactate dehydrogenase U/L 401 380 23.2 0.662 0.544 Hemoglobin g/dl 12.1 12.3 0.53 0.859 0.694 Hematocrit % 36 35.9 1.6 0.986 0.7 WBC 103/ul 15.2 16.5 1.42 0.554 0.58 RBC 106/ul 11 11.2 0.48 0.831 0.553 MCV 33.1 31.9 0.34 0.263 0.284 MCH 11 11 0.1 0.765 0.277 MCHC 33.7 34.2 0.17 0.262 0.454 Neutrophils % 76 76.7 4.5 0.843 0.46 Bands % 0.07 0.07 0.13 1 0.758 Lymphocytes % 14.3 11.4 2.65 0.105 0.108 Monocytes % 2.58 4.36 0.86 0.255 0.253 Eosinophils % 6.18 5.84 1.92 0.897 0.723 Basophils % 1.13 1.75 0.34 did not converge Fibrinogen mg/dl 244 236 11 0.658 0.64 GF = control supplement; EF = experimental su pplement; CPK = creatinine phosphokinase; GGTP = gamma glutamyl transpeptidase; WBC = white blood cell; RBC = red blood cell; MCV = mean corpuscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration

PAGE 76

64 Table 3-8. Pearson Correlati on Coefficients for correlati ons of blood glucose and BUN with kilograms of nutrients consumed and digested. --------------------Kg consumed --------------------K g digested----OM NDFOMCP NDSFSugar Starch OM* NDFOMCP* Glucose r 0.614 0.462 0.603 0.462 0.2310.495 0.560 0.400 0.595 P -value 0.020 0.097 0.023 0.097 0.4260.072 0.074 0.223 0.054 BUN r 0.427 0.146 0.419 0.146 -0.2530.816 0.521 0.202 0.583 P -value 0.128 0.619 0.136 0.619 0.382<0.001 0.100 0.552 0.060 NDFom = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber; BUN = blood urea nitrogen. *Apparent digestion Table 3-9. Pearson Correlati on Coefficients for correlati ons of blood glucose and BUN with various blood proteins. Total Total Alk protein BUN Creat CPK Albu bilirubinphos ALT AST Glucose r -0.250 0.758 -0.3030.525 0.231 -0.307 0.019 0.397 -0.078 P -value 0.389 0.002 0.2920.054 0.426 0.285 0.950 0.159 0.792 BUN r -0.190 1.000 -0.3740.515 0.212 -0.387 -0.021 0.463 0.302 P -value 0.514 0.1880.060 0.466 0.172 0.942 0.091 0.295 BUN = blood urea nitrogen; Creat = creatinine; CPK = creatinine phosphokinase; Alk phos = alkaline phosphatase; ALT = alanine aminotransferase; AST = aspartate aminotransferase

PAGE 77

65 0 100 200 300 400 500 600 700 GFEFEFGFEFEFGFEFGFGFEFGFGFEF 12123344556767 Giraffe 1Giraffe 2Giraffe 3Giraffe 4Giraffe 5Giraffe 6 treatment, period, giraffeminutes over 48 hours Total Oral Stereotypes Metal Licking Tongue Play Figure 3-1. Number of minutes over 48 hours in dividual giraffe spent engaged in specific and total oral stereotype behavior.

PAGE 78

66 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1gf1ef2ef2gf2ef3ef3gf4ef4gf5gf5ef5gf6gf6efmEq/L Figure 3-2. Blood concentrations of non-esterified fatty acids in individual giraffe. Giraffe are designated by number (1-6); gf = control diet; ef = experimental diet.

PAGE 79

APPENDIX A INDIVIDUAL ANIMAL MEASURES

PAGE 80

68Table A-1. Individu al giraffe voluntary intake (DM basis). ------Percent Body Weight -------------------Kilogr ams------------Giraffe Period Diet Total Supplement Alfalf a Hay Total Supplement Alfalfa Hay F:S Ratioa G1 1 GF 0.86 0.70 0.12 5.95 4.84 1.10 19:81 G1 2 EF 1.31 1.29 0.02 9.38 9.26 0.12 01:99 G2 1 EF 0.69 0.46 0.22 4.26 2.87 1.39 33:67 G2 2 GF 1.01 0.90 0.10 6.32 5.66 0.66 10:90 G2 3 EF 1.06 0.87 0.20 6.92 5.64 1.27 18:82 G3 3 EF 1.50 1.38 0.12 8.66 7.95 0.72 08:92 G3 4 GF 1.66 1.41 0.25 9.62 8.18 1.44 15:85 G4 4 EF 1.03 0.87 0.17 6.90 5.79 1.11 16:84 G4 5 GF 1.13 0.93 0.19 7.49 6.20 1.29 17:83 G5 5 GF 1.44 0.85 0.60 9.03 5.30 3.72 41:59 G5 6 EF 1.55 0.94 0.61 10.0 6.10 3.94 39:61 G5 7 GF 1.45 0.87 0.58 9.59 5.73 3.86 40:60 G6 6 GF 1.24 1.07 0.17 7.28 6.28 0.99 14:86 G6 7 EF 1.20 0.97 0.22 7.13 5.79 1.34 19:81 aForage:supplement intake ratio (DM basis)

PAGE 81

69 69Table A-2. Individual giraffe nutrient intake (kg) (DM basis). Giraffe Period Diet DM Ash CP NDFOMADF LigninNDSF Sugar Starch Ca P Mg K Na G1 1 GF 5.95 0.35 1.03 2.21 1.53 0.26 0.64 0.90 0.83 0.08 0.04 0.03 0.09 0.02 G1 2 EF 9.38 1.01 1.61 4.09 2.64 0.59 1.40 1.80 0.14 0.11 0.06 0.05 0.17 0.04 G2 1 EF 4.26 0.42 0.77 1.80 1.31 0.30 0.72 0.67 0.05 0.06 0.02 0.02 0.08 0.01 G2 2 GF 6.32 0.52 1.11 2.38 1.55 0.24 0.70 0.94 0.75 0.08 0.05 0.03 0.06 0.03 G2 3 EF 6.92 0.63 1.20 2.86 1.93 0.43 1.03 1.51 0.11 0.07 0.03 0.04 0.13 0.03 G3 3 EF 8.66 0.83 1.50 3.52 2.29 0.49 1.25 1.96 0.14 0.10 0.05 0.05 0.17 0.04 G3 4 GF 9.62 0.65 1.71 3.52 2.33 0.41 1.12 1.38 1.15 0.14 0.07 0.04 0.12 0.03 G4 4 EF 6.90 0.64 1.21 2.79 1.81 0.42 1.08 1.58 0.07 0.10 0.04 0.05 0.12 0.03 G4 5 GF 7.49 0.54 1.39 2.54 1.63 0.27 0.79 1.03 1.03 0.10 0.05 0.03 0.09 0.02 G5 5 GF 9.03 0.48 1.71 3.32 2.30 0.47 1.18 1.14 0.94 0.13 0.05 0.04 0.12 0.02 G5 6 EF 10.04 0.76 1.89 4.24 3.03 0.67 1.52 1.82 0.14 0.14 0.04 0.04 0.18 0.03 G5 7 GF 9.59 0.55 1.65 3.79 2.53 0.52 1.26 1.32 0.84 0.12 0.05 0.04 0.13 0.02 G6 6 GF 7.28 0.51 1.26 2.61 1.75 0.30 0.71 0.93 1.00 0.09 0.05 0.03 0.09 0.02 G6 7 EF 7.13 0.65 1.17 2.92 2.01 0.46 1.05 1.45 0.11 0.09 0.04 0.04 0.13 0.03 NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber

PAGE 82

70 70Table A-3. Individual giraffe fecal output fecal nutrient concentrati on (% of DM), and apparent nutrient digestibility (%). ----Or g anic matte r ----------NDFOM------------Crude p rotein------Giraffe Perio d Da y Diet Grams (DM)ConcentrationDi g est. ConcentrationDi g est. ConcentrationDi g est. G1 1 1 GF 222662.375.239.2 60.412.074.2 G1 1 2 GF 74369.890.746.2 84.412.691.0 G1 2 1 EF 431679.558.952.2 44.918.251.2 G1 2 2 EF 34708066.853.9 54.317.861.4 G2 1 1 EF 18136569.345.1 54.713.468.5 G2 1 2 EF 86169.184.548.9 76.713.485.1 G2 2 1 GF 269871.466.750.1 43.311.971.0 G2 2 2 GF 178374.477.152.6 60.612.779.7 G2 3 1 EF 311369.865.446.5 49.414.761.8 G2 3 2 EF 259166.572.546.5 57.913.371.3 G3 3 1 EF 310152.679.231.9 72.011.376.5 G3 3 2 EF 124564.789.742.4 85.014.987.6 G4 4 1 EF 262236.984.424.7 76.89.080.6 G4 4 2 EF 258644.181.727.8 74.29.779.3 G4 5 1 GF 265053.679.538.4 59.910.280.5 G4 5 2 GF 263053.879.636.1 62.610.480.3 G5 5 1 GF 378056.275.140.7 53.710.077.9 G5 5 2 GF 281564.878.641.7 64.610.383.0 G5 6 1 EF 409866.570.645.6 56.013.271.4 G5 6 2 EF 407858.874.140.4 61.111.674.9 G5 7 1 GF 429963.969.646.4 47.310.073.8 G5 7 2 GF 381864.372.943.7 55.99.478.1 G6 6 1 GF 249271.373.846.7 55.413.473.6 G6 6 2 GF 298157.174.936.5 58.411.173.7 G6 7 1 EF 285471.568.548.8 52.416.061.2 G6 7 2 EF 298766.169.646.9 52.114.563.0 NDFOM = neutral detergent fiber organic matter

PAGE 83

71 71Table A-4. Differences in crude protei n concentration of individua l fecal samples analyzed in wet and dried forms. Giraffe Diet Period Day Wet Dried Difference G1 GF 1 1 13.89 11.95 1.93 G1 GF 1 2 16.08 13.04 3.04 G1 EF 2 1 20.51 18.96 1.55 G1 EF 2 2 20.92 18.11 2.81 G2 EF 1 1 13.84 13.47 0.37 G2 EF 1 2 15.19 13.49 1.70 G2 GF 2 1 14.13 10.24 3.89 G2 GF 2 2 14.89 13.36 1.53 G2 EF 3 1 15.61 14.86 0.75 G2 EF 3 2 13.61 13.66 -0.06 G3 EF 3 1 13.76 12.04 1.72 G3 EF 3 2 16.36 15.23 1.13 G3 GF 4 1 15.95 13.53 2.42 G3 GF 4 2 16.54 14.97 1.56 G4 EF 4 1 10.29 6.89 3.40 G4 EF 4 2 11.76 12.63 -0.88 G4 GF 5 1 12.23 10.47 1.76 G4 GF 5 2 12.32 10.83 1.49 G5 GF 5 1 12.84 10.25 2.59 G5 GF 5 2 11.89 10.86 1.03 G5 EF 6 1 15.43 13.25 2.19 G5 EF 6 2 13.60 12.18 1.42 G5 GF 7 1 11.65 10.33 1.32 G5 GF 7 2 11.70 9.79 1.91 G6 GF 6 1 15.34 13.91 1.43 G6 GF 6 2 13.92 11.83 2.10 G6 EF 7 1 17.04 16.54 0.49 G6 EF 7 2 16.34 14.86 1.48

PAGE 84

72 72Table A-5. Minutes over 48 hour s individual giraffe spent engaged in measured behaviors. Giraffe Period Diet Sta nding Walking Lying Sleep Social contact Hay consumption Supplement consumption Total eating Eat + ruminat G1 1 GF 1421 151 1238 27 47 191 103 294 844 G1 2 EF 1582 191 1048 12 72 26 389 415 897 G2 1 EF 2001 78 775 6 22 233 173 406 1029 G2 2 GF 1889 59 876 14 46 149 114 263 719 G2 3 EF 1818 64 989 3 6 166 167 333 985 G3 3 EF 1954 444 472 11 15 153 144 297 860 G3 4 GF 1387 1035 409 0 7 218 137 355 837 G4 4 EF 1630 500 678 3 3 181 418 599 1232 G4 5 GF 1573 290 1015 23 23 324 223 547 1391 G5 5 GF 1906 12 859 38 32 599 155 754 1226 G5 6 EF 1802 87 990 26 13 484 265 749 1473 G5 7 GF 1706 105 1046 26 11 399 140 539 1207 G6 6 GF 1202 75 1622 23 12 221 92 313 1101 G6 7 EF 1357 112 1387 13 9 262 233 495 1434

PAGE 85

73 73Table A-5. Continued. Giraffe Period Diet RuminationDrinking Salt Metal licking Non-metal licking Tongue play Total oral stereotype Wall foraging Total oral G1 1 GF 550 70 0 226 2 243 471 97 1482 G1 2 EF 482 57 0 189 10 49 248 52 1254 G2 1 EF 623 5 0 510 0 9 519 0 1553 G2 2 GF 456 4 0 640 7 14 661 14 1398 G2 3 EF 652 5 0 478 1 8 487 2 1479 G3 3 EF 563 145 1 86 16 149 251 11 1268 G3 4 GF 482 184 0 33 8 168 209 7 1237 G4 4 EF 633 26 0 220 43 165 428 13 1699 G4 5 GF 844 86 0 102 75 218 395 32 1904 G5 5 GF 472 27 0 240 36 9 285 62 1600 G5 6 EF 724 38 0 222 25 25 272 49 1832 G5 7 GF 668 33 8 389 32 18 439 31 1718 G6 6 GF 788 17 0 69 12 267 348 48 1514 G6 7 EF 939 11 0 209 9 131 349 79 1873

PAGE 86

74 74Table A-6. Individual gira ffe blood values on day 21. Giraffe Period Diet NEFA Insulin GLC B UN Creat Prot Albu TBili AlkPhos ALT AST mEq/L ng/ml mg/dl mg/dl mg/d l g/dl g/dl mg/dl U/L U/L U/L G1 1 1 0.55 0.42 68 17 2.3 9.1 2.3 0.1 223 7 46 G1 2 2 0.70 0.37 92 17 2.3 9.8 2.5 0.1 238 8 44 G2 1 2 0.33 1.74 59 15 2.3 9.5 2.5 0.2 130 5 48 G2 2 1 0.78 1.05 118 22 2.1 9.3 2.5 0.1 142 8 44 G2 3 2 0.17 0.94 65 16 1.9 9.4 2.6 0.2 117 5 44 G3 3 2 0.34 0.79 105 15 1.6 8.6 2.4 0.2 114 8 37 G3 4 1 0.38 2.37 135 24 1.4 9.0 2.6 0.1 160 17 44 G4 4 2 0.25 0.83 87 17 1.6 9.5 2.5 0.2 227 8 50 G4 5 1 0.41 1.45 93 22 1.4 9.5 2.4 0.2 149 8 64 G5 5 1 0.37 0.99 89 19 1.3 7.4 2.5 0.2 117 5 47 G5 6 2 0.19 0.74 97 19 1.3 7.4 2.6 0.1 109 6 42 G5 7 1 0.22 0.84 103 21 1.5 8.2 3.0 0.1 147 11 56 G6 6 1 0.57 0.37 90 20 1.3 8.8 2.2 0.1 124 19 89 G6 7 2 0.50 0.32 64 16 1.3 9.4 2.5 0.1 109 13 50 NEFA = non-esterified fatty acids; GLC = glucose; BUN = blood ur ea nitrogen; Creat = creatinine; Prot = total protein; Albu = a lbumin; TBili = total bilirubin; AlkPhos = alkaline phosphotase; ALT = alanine ami notransferase; AST = aspertate aminotransferase

PAGE 87

75 75Table A-6 Continued. Giraffe Period Diet Ca P Na K Cl Mg Glob Lipase AmylaseTriglyc CPK GGTP Osm mg/dl mg/dl mEq/L mEq/L mEq/L mE q/L g/dl U/L U/L mg/dl U/L U/L mOsm/L G1 1 GF 7.9 9.9 147 5.1 111 2.0 6.8 37 11 52 112 20 293 G1 2 EF 7.7 11.8 147 5.0 110 2.0 7.3 27 10 34 324 11 294 G2 1 EF 8.5 8.7 147 4.6 104 1.7 7.0 28 10 20 190 18 291 G2 2 GF 8.0 12.3 145 4.1 105 1.9 6.8 16 10 37 27 13 292 G2 3 EF 8.6 11.9 145 4.9 101 2.1 6.8 27 10 34 145 17 288 G3 3 EF 9.8 12.0 149 5.1 104 1.8 6.2 31 15 37 204 12 298 G3 4 GF 9.3 15.2 148 4.7 106 1.7 6.4 26 10 31 608 9 300 G4 4 EF 8.8 10.8 145 4.6 102 2.2 7.0 21 10 58 243 18 289 G4 5 GF 8.2 6.8 145 4.1 115 2.3 7.1 38 15 51 406 24 290 G5 5 GF 9.3 9.1 144 4.6 101 2.4 4.9 37 10 35 251 46 288 G5 6 EF 9.8 10.3 146 4.9 101 2.1 4.8 78 15 29 379 25 293 G5 7 GF 9.8 11.5 156 5.0 104 3.5 5.2 55 13 42 347 27 313 G6 6 GF 8.0 8.2 143 4.2 109 1.4 6.6 30 14 33 152 15 286 G6 7 EF 9.3 10.3 146 5.1 102 2.6 6.9 66 15 39 159 19 290 Glob = globulin; Triglyc = triglycerides; CPK = creatinine phosphok inase; GGTP = gamma glutamyl transpeptidase; Osm = calculate d osmolality

PAGE 88

76 76Table A-6 Continued. Giraffe Period Diet LactD HGB HCT WBC RBC MCVMCHMCHCNeutph Bands LymphMonoctsEosinphlBasophlFibrgn U/L g/dl % 103/ul106/ul % % % % % % mg/dl G1 1 1 326 11.3 33.2 15.6 10.1 33.0 11.2 34.0 74 0 9 3 13 1 259 G1 2 2 350 12.8 37.6 16.3 11.6 32.0 11.0 34.0 65 0 15 11 7 2 218 G2 1 2 403 12.1 34.8 21.3 10.4 33.0 11.6 34.8 95 0 3 0 2 251 G2 2 1 362 11.3 33.3 14.2 9.9 34.0 11.4 33.9 76 0 16 6 2 281 G2 3 2 401 12.7 38.1 13.6 11.3 34.0 11.2 33.3 77 0 15 2 3 3 223 G3 3 2 315 12.4 37.4 15.8 10.7 35.0 11.6 33.2 77 1 6 6 10 0 200 G3 4 1 348 11.7 35.3 21.2 10.1 35.0 11.6 33.1 79 0 10 3 6 2 229 G4 4 2 325 11.0 31.6 13.9 9.8 32.0 11.3 34.8 73 0 14 2 9 2 243 G4 5 1 352 14.0 40.9 13.0 12.7 32.0 11.0 34.2 83 0 11 1 4 1 242 G5 5 1 442 13.2 38.5 15.5 12.3 31.3 10.7 34.3 61 24 1 12 2 242 G5 6 2 398 12.8 37.7 14.1 12.1 31.0 10.6 34.0 65 0 19 7 6 3 293 G5 7 1 462 14.3 42.2 12.3 13.8 31.0 10.4 33.9 68 0 25 1 5 1 248 G6 6 1 530 10.4 30.8 17.9 10.0 31.0 10.4 33.8 85 0 11 2 2 0 236 G6 7 2 448 11.1 32.6 16.5 10.9 30.0 10.2 34.0 84 0 12 2 2 0 215 LactD = lactate dehydrogenase; HGB = hemoglobin; HCT = hemato crit; WBC = white blood cell; RBC = red blood cell; MCV = mean cor puscular volume; MCH = mean corpuscular hemoglobin; MCHC = mean corpuscular hemoglobin concentration

PAGE 89

77 77Table A-7. Body weight and body condition score of i ndividual giraffe on days 1 and 21 of each period. ---------------Body weight (kg)--------------Body condition score Giraffe Period Treatment Day 1 Day 28 Change Day 1 Day 28 G1 1 GF 693 694 1 4.0 4.0 G1 2 EF 694 717 23 4.0 4.5 G2 1 EF 625 622 -4 3.5 3.5 G2 2 GF 622 630 8 3.5 3.5 G2 3 EF 630 652 23 3.5 3.5 G3 3 EF 542 559 17 3.5 4.5 G3 4 GF 577 580 2 5.0 4.5 G4 4 EF 673 668 -5 4.0 4.0 G4 5 GF 668 667 -1 4.0 4.0 G5 5 GF 611 626 15 3.5 4.0 G5 6 EF 626 651 25 4.0 4.5 G5 7 GF 651 661 10 4.5 4.5 G6 6 GF 590 589 -1 2.5 3.0 G6 7 EF 589 598 10 3.0 4.0

PAGE 90

78 78 % of BW 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 GFEFEFGFEFEFGFEFGFGFEFGFGFEF 12123344556767 G1G2G3G4G5G6 Alfalfa Hay Supplement Figure A-1. Individual animal cons umption of alfalfa hay and supplemen t as a percentage of body weight. Giraffe are designated by number (G1-G6); GF = control diet; EF = experimental diet

PAGE 91

79 79 kg 0 1 2 3 4 5 6 7 8 GFEFEFGFEFGFGFEFGFGFEF 12123556767 G 1G 2G 4G 5G 6 treatment, period, giraffe OM NDFOM CP Figure A-2. Individual animal di gestion of neutral detergent fibe r organic matter (NDFOM), and a pparent digestion of OM and CP Giraffe are designated by number (G1-G6); GF = control diet; EF = experimental diet

PAGE 92

80 APPENDIX B CARBOHYDRATE FRACTIONING IN FEEDSTUFFS Figure B-1. Carbohydrate part itioning. (Hall, 2001) Table B-1. NDF and NFC fractions (percent of sample DM) in feedstuffs analyzed at the University of Florida. Feedstuff NDF Sugars Starch Soluble Fiber Alfalfa hay 37.4 10.0 3.0 16.2 Citrus pulp 22.1 26.5 1.0 32.9 Corn meal 16.0 2.3 62.6 8.5 Cottonseed hulls <1 4.0 Soybean meal (48%) 12.8 11.2 1.5 15.9 Sugar beet pulp 44.6 12.8 0.0 30.0 Ground wheat 12.1 1.8 64.6 8.8 Data means from Hall, 2001. Plant CarbohydratesCell Contents Cell WallHemicelluloses Pectic Substances -glucans Fructans Starches Mono+Oligosaccharides Organic Acids Cellulose ADF NDSF NDSCGalactans Non-Starch Polysaccharides NDF

PAGE 93

81 APPENDIX C INFORMATION ON CONTROL DIET COMPOSITION AND BEHAVIOR RECORDING Table C-1. Mean analyzed chemical compos ition of the batches of Purina Omelene 200 (n=5) and Mazuri Browser Breeder (n =5) fed to captive giraffe during a giraffe feeding study. Nutrient Purina Omelene 200 Mazuri Browser Breeder DM, % 85.4 85.4 OM, % 93.6 93.6 Ash, % 6.37 6.37 CP, % 16.6 16.6 NDFOM, % 17.2 17.2 ADF, % 7.47 7.47 Lignin, % 1.84 1.84 NDSF, % 6.93 6.93 Sugar, % 11.0 11.0 Starch, % 38.1 38.1 Ca, % 0.68 0.68 P, % 0.56 0.56 Mg, % 0.20 0.20 K, % 1.15 1.15 Na, % 0.23 0.23 Fe, ppm 390 297 Mn, ppm 217 115 Zn, ppm 222 137 Cu, ppm 45.6 16.8 NDFOM = neutral detergent fiber organic matter; NDSF = neutral detergent soluble fiber Purina Omelene 200 is a sweet feed containing: whol e oats, cracked corn, ground corn, dehulled soybean meal, corn flour, cane molasses, soybean oil, wheat middlings, dicalcium phosphate, sodium selenite, propionic acid, dried whey, vitamin E supplement, choline chloride, citric acid, vitamin A supplement, calcium pantothenate, ferric oxide, tocophorols, riboflavin supplement, vitamin B-12 supplement, vitamin D3 supplement, niacin supplement, ferrous carbonate, manganous oxide, zinc oxide, copper sulfate, magnesium oxide, ferrous oxide, calcium iodate, cobalt carbonate, DL-methionine, L-lysine. Mazuri Browser Breeder is a pelleted concentrate containing: ground soybean hulls, wheat middlings, ground aspen, dehulled soybean meal, dried beet pulp, cane molasses, dehydrated alfalfa meal, sucrose, brewers dried yeast, soybean oil, dicalcium phosphate, salt, calcium carbonate, magnesium oxide, menadione dimethylpyrimidinol bisulfite (source of vitamin K), pyridoxine hydrochloride, dl-alpha tocopheryl acetate (source of vitamin E), cholecalciferol (source of vitamin D3), vitamin A acetate, calcium pantothenate, ethoxyquin (a preservative), thiamin mononitrate, cyanocobalamin (source of vitamin B12), riboflavin, biotin, nicotinic acid, manganous oxide, zinc oxide, ferrous carbonate, copper sulfate, zinc sulfate, calcium iodate, cobalt carbonate, sodium selenite.

PAGE 94

82 Table C-2. Behavioral categor y definitions used by observe rs when recording giraffe behaviors. Cud Chewing cud (ruminating) Supp Eating supplement from feed pan Hay Eating hay from hay rack Drink Drinking Salt Licking salt block Dirt Eating dirt Metal Licking metal Lick Licking or mounthing non-metal inanimate objects Wall Foraging in or mouthing area behind back wall Tongue Tongue playing not associated with licking, eating, drinking, or ruminating Social Making physical contact with other giraffe Stand Standing Walk Walking Lay Lying down with head up Sleep Sleeping lying down with head folded over onto flank

PAGE 95

83 Figure C-1. Example of data sheet used to record giraffe behavior. Giraffe Behavior LogOctob. 14, 020:00-1:00UF/ BGT Hoofstock Feeding Study C u d S upp H ay W ater S a l t Di rt M eta l Li c k W a ll T ongue S oc i a l S tan d W a lk L ay Sl eep 0:00 0:01 0:02 0:03 0 : 04 0:05 0:06 0:07 0:08 0 : 09 0:10 0:11 0:12 0:13 0 : 14 0:15 0:16 0:17 0:18 0 : 19 0:20 0:21 0:22 0:23 0 : 24 0:25 0:26 0:27 0:28 0 : 29 0:30 0:31 0:32 0:33 0 : 34 0:35 0:36 0:37 0:38 0 : 39 0:40 0:41 0:42 0:43 0 : 44 0:45 0:46 0:47 0:48 0 : 49 0:50 0:51 0:52 0:53 0 : 54 0:55 0:56 0:57 0:58 0 : 59 Left Pen = grey Observer:Right Pen = white

PAGE 96

84 LIST OF REFERENCES Allen MS. 1996. Physical constraints on volunt ary intake of forages by ruminants. J Anim Sci 74:3063-3075. Allen MS. 1997. Relationship between fermenta tion acid production in the rumen and the requirement for physically effec tive fiber. J Dairy Sci 80:1447-1462. Ariza P, Bach A, Stern MD, Hall MB. 2001. E ffects of carbohydrates from citrus pulp and hominy feed on microbial fermentati on in continuous culture. J Anim Sci 79:2713-2718. Association of Official Anal ytical Chemists. 1990. Official Methods of Analysis, 15th edition. Arlington, VA. Baer DJ, Oftedal OT, Fahey Jr. GC. 1985. F eed selection and digestibility by captive giraffe. Zoo Biol 4:57-64. Bailey CB. 1961. Saliva secretion and its relation to feeding in cattle. III. The rate of secretion of mixed saliva in the cow during eating, with an estimate of the magnitude of the total daily secretion of mixed saliva. Br J Nutr 15:443-451. Ball RL, Kearney C, Burton M, Dumoneux G, Olsen JH. 2002. Morbidity and mortality related to hypoglycemia and chronic energy malnutrition in captive giraffe. In: Proc Am Assoc Zoo Vets, p 180-185. Barnes R, Greene K, Lamm M. 2002. Manageme nt and husbandry of duikers at the Los Angeles Zoo. Zoo Biol 21:107-121. Bashaw MJ, Tarou LR, Maki TS, Maple TL. 2001. A survey assesment of variable related to stereotypy in captive gira ffe. Appl Anim Behaviour Sci 73:235-247. Beauchemin KA, Yang WZ, Rode LM. 2003. Eff ects of particle size of alfalfa-based dairy cow diets on chewing activity, rumi nal fermentation, and milk production. J Dairy Sci 86:630-643. Beharka AA, Nagaraja TG, Morrill JL, Ke nnedy GA, Klemm RD. 1998. Effects of form of diet on anatomical, microbial, and fe rmentative development of the rumen of neonatal calves. J Dairy Sci 81:1946-1955.

PAGE 97

85 Ben-Ghedalia D, Yosef E, Mi ron J, Est Y. 1989. The effects of starchand pectin-rich diets on quantitative aspects of digestion in sheep. An im Feed Sci Technol 24:289298. Bernard L, Chaise JP, Baumont R, Poncet C. 2000. The effect of physical form of orchardgrass hay on the passa ge of particulate matter th rough the rumen of sheep. J Anim Sci 78:1338-1354. Bhattacharya AN, Lubbadah WF. 1971. Feed ing high levels of beet pulp in high concentrate dairy rati ons. J Dairy Sci 54:95-99. Burton MS, Dierenfeld ES. 1990. Comparisons of vitamin E injectables in giraffes (Giraffa Camelopardalis reticulata). In: Proc Am Assoc Zoo Vets, p 198-201. Caister LE, Shields WM, Gosser A. 2003. Female tannin avoidance: a possible explanation for habitat and diet ary segregatoin of giraffes (Giraffa camelopardalis peralta) in Niger. Afr J Ecol 41:201-210. Callison SL, Firkins JL, Eastridge ML, Hull BL. 2001. Site of nutrient digestion by dairy cows fed corn of different particle si zes or steam-rolled. J Dairy Sci 84:1458-1467. Campbell CP, Marshall SA, Mandell IB, W ilton JW. 1992. Effects of source of dietary neutral detergent fiber on chewing behavior in beef cattle fed pelleted concentrates with or without supplementa l roughage. J Anim Sci 70:894-903. Campling RC, Balch CC. 1961. Factors affecting the voluntary intake of food by cows. 1. Preliminary observations on the effect, on th e voluntary intake of hay, of changes in the amount of the reticulo-rumi nal contents. Br J Nutr 15:523-530. Cassida KA, Stokes MR. 1986. Eating and restin g salivation in early lactation dairy cows. J Dairy Sci 69:1282-1292. Chaffe PS. 1968. Report on death of a giraffe. J Small Anim Pract 9:136-137. Chai K, Kennedy PM, Milligan LP. 1984. Reducti on in particle size during rumination in cattle. Can J Anim Sci 64(Suppl.):339-340. Chester-Jones H, Metwally HM, Linn JG Ziegler DM, Stern MD. 1991. Effects of dietary protein-energy interr elationships on Holstein steer performance and ruminal bacterial fermentation in conti nuous culture. J Anim Sci 69:4956-4966. Ciofolo I, LePendu Y. 2002. The feeding beha viour of giraffe in Niger. Mammalia 66:183-194. Clark PW, Armentano LE. 2002. Influence of pa rticle size on the effectiveness of the fiber in alfalfa silage. J Dairy Sci 85:3000-3007.

PAGE 98

86 Clauss M. 1998. Feeding Giraffe (Giraffe camelopardis) [Masters thesis]. London, UK: University of London. 114 p. Clauss M, Kienzle E, Wiesner H. 2002a. Impor tance of the wasting syndrome complex in captive moose (Alces alces). Zoo Biol 21:499-506. Clauss M, Lechner-Doll M. 2001. Differences in selective reticul o-ruminal particle retention as a key factor in rumina nt diversification. Oecologia 129:321-327. Clauss M, Lechner-Doll M, Flach EJ, Tack C, Hatt JM. 2001. Comparative use of four different marker systems for the estimation of digestibility and low food intake in a group of captive giraffes (Giraffa camelopardalis). Zoo Biol 20:315-329. Clauss M, Lechner-Doll M, Flach EJ, Wi sser J, Hatt J-M. 2002b. Digestive tract pathology of captive giraffe (Giraffa camelopardalis) an unifying hypothesis. European Association of Zoo and Wildlife Veterinarians Fourth Scientific Meeting. Heidelberg, Germany. p 99-107. Clemens ET, Maloiy GMO, Sutton JD. 1983. Mola r proportions of volatile fatty acids in the gastrointestinal tract of East African wild ruminants. Comp Biochem Physiol 76A:217-224. Dagg AI, Foster JB. 1976. The Giraffe. Its Bi ology, Behavior, and Ecology. Malabar, FL: Krieger Publishing Company. Dehority BA, Odenyo AA. 2003. Influence of diet on the rumen protozoal fauna of indigenous African wild ruminants. J Eukaryot Microbiol 50:220-223. Dierenfeld ES, Mueller PJ, Hall MB. 2002. Duikers: Native food composition, micronutrient assessment, and implications for improving captive diets. Zoo Biol 21:185-196. Dierenfeld ES, Wildman RE C, Romo S. 2000. Feed inta ke, diet utilization, and composition of browses consumed by the Sumatran rhino (Dicerorhinus sumatrensis) in a North American zoo. Zoo Biol 19:169-180. Dirksen GU, Liebich HG, Mayer E. 1985. Adaptive changes of the ruminal mucosa and their functional and clinical significa nce. The Bovine Practitioner 20:116-120. Ditchkoff SS. 2000. A decade since "diversifi cation of ruminants": has our knowledge improved? Oecologia 125:82-84. Enqvist KE. 2003. Dental disease and serous atrophy of fat syndrome in captive giraffes (Giraffa camelopardalis). Minneapolis, MN: Annual c onference of the American Association of Zoo Veterinarians. Field CR, Ross IC. 1976. The savanna ecology of Kidepo Valley National Forest. E Afr Wildl J 14:1-15.

PAGE 99

87 Firkins JL. 1997. Effects of feeding nonforage fi ber sources on site of fiber digestion. J Dairy Sci 80:1426-1437. Flach EJ, Thornton SM, Kirkwood JK, Sain esbury AW. 1997. Chronic loss of condition with persistent neutrophilia in a reticulated giraffe (Giraffa camelopardalis). In: Proc British Vet Zoo Soc, p 33-37. Foose TJ. 1982. Trophic strategies of rumina nt versus nonruminant ungulates [PhD Dissertation]. Chicago, IL: The University of Chicago. 334 p. Foster JB. 1966. The giraffe of Nairobi Nati onal Park: home range, sex ratios, the herd and food. E Afr Wildl J 4:139-148. Fowler ME. 1978. Peracute mortality in cap tive giraffe. J Am Vet Med Assoc 173:10881093. Fox H. 1938. The giraffe. Some notes upon the natu ral characters of th is animals, its care and its misfortune. In: Zso, editor. Repor t of the Penrose Research Laboratory, formerly the Laboratory and Museum of Comparative Pathology. Philadelphia, PA: Zoological Society of Philadelphia. p 35-67. Friggens N, Emmans GC, Robertson S, Chamberlain DG, Whittemore CT, Oldham JD. 1995. The lactational responses of dairy cows to amount of feed and to the source of carbohydrate energy. J Dairy Sci 78:1734-1744. Friggens NC, Oldham JD, Dewhurst RJ, Horg an G. 1998. Proportions of volatile fatty acids in relation to the chemical compositi on of feeds based on grass silage. J Dairy Sci 81:1331-1344. Furstenburg D, Van Hoven W. 1994. Condensed ta nnin as an anti-defo liate agent against browsing by giraffe (Giraffe camelopardalis) in the Kruger National Park. Comp Biochem Physiol 107A(2):425-431. Garrett EF, Pereira MN, Nordlund KV, Arme ntano LE, Goodger WJ, Oetzel GR. 1999. Diagnostic methods for the det ection of subacute ruminal acidosis in dairy cows. J Dairy Sci 82:1170-1178. Glock RD, DeGroot BD. 1998. Su dden death of feedlot cattle. J Anim Sci 76:315-319. Goering HK, Van Soest PJ. 1970. Forage fiber an alysis (apparatus, reagents, procedures and some applications). In: Agric. Handb. No. 379. Washington, DC: ARS-USDA. Gordon IJ, Illius AW. 1996. The nutritiona l ecology of African ruminants: a reinterpretation. J Anim Ecol 65:18-28. Grant RJ. 1997. Interactions among forages a nd nonforage fiber sources. J Dairy Sci 80:1438-1446.

PAGE 100

88 Hall MB. 1999. Management strategies agai nst ruminal acidosis. In: Tenth Annual Florida Ruminant Nutrition Sy mposium. Gainesville, FL: University of Florida. p 104-113. Hall MB. 2001. Neutral Detergent-Soluble Ca rbohydrates, Nutriti onal Relevance and Analysis, A Laboratory Manual. Gainesvelle: University of Florida Report nr 339. Hall MB, Herejk C. 2001. Differences in yields of microbial crude protein from in vitro fermentation of carbohydrates. J Dairy Sci 84:2486-2493. Hall MB, Hoover WH, Jennings JP, Miller Webster TK. 1999. A method for partitioning neutral detergent-soluble carbohydrat es. J Sci Food Agric 79:2079-2086. Hall-Martin AJ, Skinner JD, Va n Dyk JM. 1975. Reproduction in the giraffe in relation to some environmental factors. E Afr Wildl J 13:237-248. Hatt JM, Lechner-Doll M, Mayes B. 1998. The use of dosed and herbage n-alkanes as markers for the determination of digestive strategies of captive giraffes (Giraffa camelopardalis). Zoo Biol 17:295-309. Heldt JS, Cochran RC, Stokka GL, Farmer CG, Mathis CP, Titgemeyer EC, Nagaraja TG. 1999. Effects of different supplemental sugars and starch fed in combination with degradable intake protein on low-quali ty forage use by beef steers. J Anim Sci 77:2793-2802. Hofmann RR. 1973. The Ruminant Stomach--Stomach Structur e and Feeding Habits of East African Game Ruminants. Odhiambo TR editor. Nairobi, Kenya: East African Literature Bureau. Hofmann RR. 1984. Comparative anatomical studies imply adaptive variations of ruminants digestive physiology. J Anim Sci 64:203-205. Hofmann RR. 1988. Morphophysiological evoluti onary adaptations of the ruminant digestive system. In: Dobson A, Dobson MJ, editors. Aspects of Digestive Physiology in Ruminants: Proceedings of a Satellite Symposium of the 30th International Congress of th e International Union of P hysiological Sciences. p 120. Hofmann RR, Kock RA, Ludwig J, Axm acher H. 1988. Seasonal changes in rumen papillary development and body condition in free ranging Chinese water deer (Hydropotes inermis). J Zool 216:103-117. Hofmann RR, Matern B. 1988. Changes in gastrointestinal morphology related to nutrition in giraffes. C ons Sci and Zoos 22:168-176. Horwitz W. 1982. Evaluation of analytical methods used for regulation of food and drugs. Anal Chem 54:67A-76A.

PAGE 101

89 Huntington GB. 1997. Starch utilization by rumi nants: from basics to the bunk. J Anim Sci 75:852-867. Johnson MM, Peters JP. 1993. Technical no te: An improved method to quantify nonesterified fatty acids in bovine plasma. J Anim Sci 71:753-756. Jorgenson NA, Schultz LH. 1963. Ration effects on rumen acids, ketogenesis, and milk composition. I. Unrestricted rougha ge feeding. J Dairy Sci 46:437-443. Junge RE, Bradley TE. 1993. Peracute morta lity syndrome of giraffes. In: Fowler ME, editor. Zoo and Wild Animal Medicine : Current Therapy 3. Philadelphia, PA: Saunders Company. p 547-549. Kay RND, Engelhardt WV, White RG. 1980. The digestive physiology of wild ruminants. In: Ruckebusch Y, Thive nd P, editors. Digestive Physiology and Metabolism in Ruminants. Wesport, CT: AVI. Kennedy PM, McSweeney CS, Welch JG. 1992. In fluence of dietary particle size on intake, digestion, and passage rate of digesta in goats and sheep fed wheaten (Triticum aestivum) hay. Small Ruminant Research 9:125-138. Kononoff PJ, Heinrichs AJ. 2003. The effect of corn silage particle size and cottonseed hulls on cows in early lactation. J Dairy Sci 86:2438-2451. Krajcarski-Hunt H, Plaizier JC, Walton J-P, Spratt R, McBride BW. 2002. Short communication: Effect of subacute rumina l acidosis on in situ fiber digestion in lactating dairy cows. J Dairy Sci 85:570-573. Krause KM, Combs DK. 2003. Effects of forage particle size, forage source, and grain fermentability on performance and ruminal pH in midlactation cows. J Dairy Sci 86:1382-1397. Krause KM, Combs DK, Beauchemin KA. 2002. E ffects of forage particle size and grain fermentability in midlactation cows. II. Ruminal pH and chewing activity. J Dairy Sci 85:1947-1957. Lackey LB, LaRue FV. 1997. North American Regional Giraffe Studbook. Dallas, TX: Dallas Zoo. Lechowski R, Pisarski J, Goslawski J, Lenarcik M. 1991. Exocrine pancreatic insufficiency-like syndrome in gira ffe. J Wildlife Diseases 27:728-780. Leuthold BM, Leuthold W. 1972. F ood habits of giraffe in Tsavo National Park, Kenya. E Afr Wildl J 10:129-141. Leuthold BM, Leuthold W. 1978. Daytime activ ity patterns of gere nuk and giraffe in Tsavo National Park, Kenya. E Afr Wildl J 16:231-243.

PAGE 102

90 Lintzenich BA, Ward AM. 1997. Hay and pellet ratios: considerations in feeding ungulates. Nutrition Advisory Group Handbook: Fact Sheet 007. www.nagonline.net. Last accessed: December 29, 2004. Maekawa M, Beauchemin KA, Christensen DA. 2002. Effect of concentrate level and feeding management on chewing activities saliva production, and ruminal pH of lactating dairy cows. J Dairy Sci 85:1165-1175. Martin P, Bateson P. 1986. Measuring Beha vior: An Introductory Guide. New York: Cambridge Univ. Press. Martin SA, Sullivan HM, Evans JD. 2000. Effect of sugars and malate on ruminal microorganisms. J Dairy Sci 83:2574-2579. McSweeney CS, Kennedy PM. 1992. Influence of dietary particle size on chewing activity and reticulo-ruminal motility in goats and sheep fed wheaten (Triticum aestivum) hay. Small Ruminant Research 9:107-115. Mertens DR. 1997. Creating a system for meeti ng the fiber requirements for dairy cows. J Dairy Sci 80:1463-1481. Miller M, Abou-Madi N, Coville B, Olsen J. 1996. Comparison of in vitro tests for evaluation of passive transfer in gira ffe. In: Proc Am Assoc Zoo Vets, p 94-101. Miller WJ, O'Dell GD. 1969. Nutritional pr oblems of using maximum forage or maximum concentrates in dairy rations. J Dairy Sci 52:1144-1154. Moore JA, Poore MH, Swingle RS. 1990. Influen ce of roughage source on kinetics of digestion and passage, and on calculated extents of rumina l digestion in beef steers fed 65% concentrate diets. J Anim Sci 68:3412-3420. Nocek JE. 1997. Bovine acidosis: implicati ons on laminitis. J Dairy Sci 80:1005-1028. Owens FN, Goetsch AL. 1988. Ruminal ferm entation. In: Church DC, editor. The Ruminant Animal Digestive Physiology and Nutrition. Prospect Heights, IL: Waveland Press, Inc. Owens FN, Secrist DS, Hill WJ, Gill DR. 1998. Ac idosis in cattle: a review. J Anim Sci 76:275-286. Paglia DE, Miller RE. 1992. Erythrocyte ATP deficiency and acatalasemia in the black rhinoceros (Diceros bicornis) and their pathogenetic ro les in acute episodic hemolysis and mucocutaneous ulcerations In: Proc Am Assoc Zoo Vets, p 217219. Pasha TN, Prigge EC, Russell RW, Bryan WB 1994. Influence of moisture content of forage diets on intake and digest ion by sheep. J Anim Sci 72:2455-2463.

PAGE 103

91 Pellew RA. 1983. The giraffe and its food res ource in the Serengeti. I. Composition, biomass and production of availabl e browse. Afr J Ecol 21:241-267. Pellew RA. 1984. Food consumption and energy budgets of the giraffe. J Appl Ecol 21:141-159. Poppi DP, Norton BW, Minson DJ, Hendricksen RE. 1980. The validity of the critical size theory for particles leaving the rumen. J Agric Sci 94:275-280. Randhawa SS, Gupta PP, Misra SK. 1981. Hist opathological changes in experimental ruminal acidosis in buffalo calves. Ind J Anim Sci 51:518-521. Rowell-Schafer A, Dyer J, Hofmann RR, L echner-Doll M, Meyer HHD, Shirazi-Beechey SP, Streich WJ. 1999. Abundance of intestinal Na+/glucose cotransporter (SGLT1) in roe deer (Capreolus capreolus). J Anim Physiol 82:25-32. Russell JB, Wilson DB. 1996. Why are ruminal ce llulolytic bacteria unable to digest cellulose at low pH? J Dairy Sci 79:1503-1509. SAS. 1999. The SAS System for Windows, Rel ease 8.02. Cary, NC: SAS Institute, Inc. Schettini MA, Prigge EC, Nestor EL. 1999. Influence of mass and volume of ruminal contents on voluntary intake a nd digesta passage of a forage diet in steers. J Anim Sci 77:1896-1904. Shain DH, Stock RA, Klopfenstein TJ, Herold DW. 1999. The effect of forage source and particle size on finishing yearling steer performance and ruminal metabolism. J Anim Sci 77:1082-1092. Shipley L. 1999. Grazers and browsers: how di gestive morphology affects diet selection. In: Launchbaugh KL, Sanders KD, Mosley JC, editors. Grazing Behavior of Livestock and Wildlife. Moscow, ID: University of Idaho, p 20-27. Shochat E, Robbins CT, Parish SM, Young PB, Stephenson TR, Tamayo A. 1997. Nutritional investigations and manageme nt of captive moose. Zoo Biol 16:479-494. Smith HA, Jones TC, Hunt RD. 1972. Veteri nary Pathology. Philadelphia, PA: Lea & Febiger. Sniffen CJ, O'Connor JD, Van Soest PJ, Fox DG, Russell JB. 1992. A net carbohydrate and protein system for evaluating cattl e diets: II. Car bohydrate and protein availability. J Anim Sci 70:3562-3577. Soeldner JS, Sloane D. 1965. Critical variable s in the radioimmunoassay of serum insulin using the double antibody t echnic. Diabetes 14:711-719. Stone WC. 2004. Nutritional approaches to minimize subacute ruminal acidosis and laminitis in dairy cattle. J Dairy Sci 87(E. Suppl.):E13-E26.

PAGE 104

92 Strafuss AC, Kennedy GA. 1973. Degenerative myopathy in a giraffe. J Am Vet Med Assoc 163:551-552. Strandberg J, Eckhaus M, Kincaid A, Cr anfield M. 1984 Fatal wasting disease in Angolan giraffes. In: Proc Am Assoc Zoo Vets, p 115-116. Strobel HJ, Russell JB. 1986. Effect of pH and energy spilling on bacterial protein synthesis by carbohydrate-limited cultures of mixed rumen bacteria. J Dairy Sci 69:2941-2947. Swenson MJ. 1984. Dukes' Physiology of Dome stic Animals, 10th edition. Ithaca, NY: Cornell University Press. Tarou LR, Bashaw MJ, Maple TL. 2003. Failure of a chemical spray to significantly reduce stereotypic licking in a captive giraffe. Zoo Biol 22:601-607. Theurer CB, Swingle RS, Wanderley RC, Kattnig RM, Urias A, Ghenniwa G. 1999. Sorghum grain flake density and source of r oughage in feedlot ca ttle diets. J Anim Sci 77:1066-1073. Van Soest PJ. 1994. Nutritional Ecology of th e Ruminant. 2nd ed. Ithaca, NY: Cornell University Press. 476 p. Van Soest PJ, Robertson JB. 1985. Analysis of Forages and Fibrous Foods--A Laboratory Manual for Animal Science 613. Ithaca, NY: Cornell University. 202 p. Van Soest PJ, Robertson JB, Lewis BA. 1991. Methods for dietar y fiber, neutral detergent fiber, and nonstarch polysacchar ides in relation to animal nutrition. J Dairy Sci 74:3583-3597. Van Vuuren AM, Van Der Koelen CJ, Vroons-De Bruin J. 1993. Ryegrass versus corn starch or beet pulp fiber diet effects on digestion and inte stinal amino acids in dairy cows. J Dairy Sci 76:2692-2700. Weston RH, Cantle JA. 1984. The movement of undigested plant particle fractions through the stomach of r oughage-fed young sheep. Can J Anim Sci 64(Suppl.):322323. Willette MM, Norton TL, Miller CL, Lamm MG. 2002. Verterinary concerns of captive duikers. Zoo Biol 21:197-207. Wolfe BA, Sladky KK, Loomis MR. 2000. Obstructiv e urolithiasis in a reticulated giraffe (Giraffe camelopardalis reticulata). The Vet Rec 146:260-261. Wood S, Dyer J, Hofmann RR, Shirazi-Beechey SP. 2000. Expression of the Na+/glucose co-transporter (SGLT1) in the intestine of domestic and wild ruminants. Pflugers Arch 441:155-162.

PAGE 105

93 Yang WZ, Beauchemin KA, Rode LM. 2001. Eff ects of grain processing, forage to concentrate ratio, and forage particle size on rumen pH and digestion by dairy cows. J Dairy Sci 84:2203-2216.

PAGE 106

94 BIOGRAPHICAL SKETCH Celeste Kearney grew up in the small fi shing and farming community of Oquawka, IL. She worked in the equine industry for se veral years, and completed a B.S. in animal sciences at Murray State Univ ersity in 1997. After working in the zoo industry for four years, she enrolled in graduate studies in an imal sciences at the University of Florida.