lon bags in a washing machine. Func- tional specific gravity was estimated in a pycnometer for unwashed and washed samples after 0 and 15 h of soaking in.
NUTRITION, FEEDING, AND CALVES
Solubility, Water-Holding Capacity, and Specific Gravity of Different Concentrates1 MAURIZIO RAMANZIN, LUCIA BAILONI, and GIOVANNI BITTANTE Dipartimento di Scienze Zootecniche Universita di Padova 35131 Padova. Italy ABSTRACT
(Key words: water-holding capacity, functional specific gravity, unit specific gravity, solubility, concentrates)
The effects of solubility and waterholding capacity on functional and unit specific gravity were evaluated in samples of fish. soybean, linseed, corn gluten. corn gluten feed, corn. barley. and dehydrated alfalfa meals and in wheat bran. raw flaked soybeans, cottonseeds. and dried sugar beet pulp. Solubility was estimated by washing the samples in nylon bags in a washing machine. Functional specific gravity was estimated in a pycnometer for unwashed and washed samples after 0 and 15 h of soaking in distilled water. Water-holding capacity was measured by a centrifugation method and by a filtration method. Unit specific gravity was estimated as the weighted mean of the functional specific gravity of insoluble DM and the specific gravity of water held by the particles. Solubility and functional specific gravity of insoluble DM varied significantly among feedstuffs from 5.0 to 53.2% of DM and from 1.31 to 1.62. respectively. The increase in functional specific gravity from soaking was small. Waterholding capacity was lower with filtration than with centrifugation methods and varied from .94 to 6.44 g of H20/g of OM. Unit specific gravity varied significantly from 1.07 to 1.24. Soluble fractions and water-holding capacity can markedly influence the functional and unit specific gravity of concentrate particles.
Abbreviation key: FSG = functional SO, GV = gas-filled voids. SG = specific gravity, USG = unit sa, WHC = water-holding capacity. WL = washing losses. INTRODUCTION
Received May 17. 1993. Accepted October 18. 1993. IResearch supported by Ministero dell'Universitil e Ricerca Scientifica e Tecnologica (40%). Project Valutazione energetica e proteica di diete per ruminanti. 1994 J Dairy Sci 77:774-781
Knowledge of rumen passage rate of feed particles is important in order to develop reliable models of rumen fermentation. To thiS purpose, feed DM can be divided into soluble and insoluble fractions. Soluble DM is assumed to be almost instantaneously degraded (17), and, in any case, its rumen passage rate is that of the liquid phase, which is easily measurable (5). Insoluble OM can be degraded at different rates, and its passage rate can vary widely, depending on physical and chemical properties of feed particles (4, 13, 19). A major factor influencing passage rate of particulate matter is specific gravity (SG) (5, 13). Because soluble OM may differ in sa from insoluble DM (28, 29), solubility may influence SO of feed particles. In addition, the sa of feed particles can be modified by the gas and liquid associated with the solid (26, 27, 28). The terminology reported by Wattiaux et al. (27) defines as functional sa (FSG) the SO of the solid and gas fractions and defines as unit sa (USG) the SO of the solid, gas, and liquid fractions of digesta particles. In forage particles, FSa increases with hydration because of replacement of entrapped gas by the liquid. Kinetics of the increase are affected by type of forage, particle size, and solution used (11, 12, 27). Water uptake by feed particles may not only replace entrapped gas but also saturate the insoluble DM and cause a swelling of the particles. The
774
775
WATER-HOLDING CAPACITY AND SPECIFIC GRAVITY
water held by a swollen particle modifies its actual weight and volume. Therefore, differences in water-holding capacity (WHC) influence the USG of feed particles. The objectives of this experiment were to compare the effects of solubility and WHC on FSG and USG of different concentrates and to investigate the relationships between these parameters and chemical composition. MATERIALS AND METHODS Feedstuffs and Treatments
The 12 energy and protein concentrates used in this study are listed in Table 1. The feedstuffs chosen were among those most commonly used and were selected in order to obtain a wide range in chemical composition. Chemical characteristics, analyzed according to AOAC (2) and Goering and Van Soest (9), also are given in Table 1. All feedstuffs were ground in a laboratory hammer mill with a screen diameter of 2 mm. Samples of the ground feedstuffs were placed into tared polyester bags (10 x 15 em, 40-~m pore size; Gaudenzi, Padova, Italy) in an amount equal to 15 mg of DM/cm 2 of inner free surface of the bag. The bags were washed in a washing machine (REX RF421; Zanussi, Pordenone, Italy) for 30 min with water at 30'C. The washing cycle consisted of 15 min of washing and 4 min of centrifugation at a calculated 40 x g. The bags then were dried in a ventilated oven for 48 h at 60'C, and samples were used for SG and WHC determinations. Measurements
Soluble fractions of OM were measured as washing losses (WL) of OM from two polyester bags for each feedstuff. A preliminary experiment showed that fine material passing through a sieve with a pore size of 38 ~m (Vibro laboratory sieving machine; Retsch, Haan, Germany) was less than 1% of OM in all the feedstuffs. The FSG of OM was measured for both the unwashed and washed feedstuffs according to Hooper and Welch (12). Samples of each feedstuff (2.0 ± .05 g) were put into Gay-Lussac pycnometers (Vetrotecnica, Venezia, Italy) in 50 ml of distilled water at 20'C, and .002 ml of Combiotic (Pfizer Co.,
New York, NY) were added. The FSG was measured after 0 and 15 h of soaking with four pycnometers (two replicates and two soaking times). The FSG of soluble OM was calculated as the weighted difference from the FSG of unwashed and washed OM. Estimates of hydration in pycnometers were assumed to be estimates of the water that replaced gas-filled voids (gas volume; GV) because the water required to saturate the insoluble OM is much smaller (27). Estimates were obtained as the difference between the volumes of distilled water displaced by feed particles in the pycnometers after 15 h of soaking and those displaced after 0 h (27), using the following equation: GV
= lIFSG 15
-
IIFSG o
where FSG 15 and FSGo are FSG values measured after 15 and 0 h of soaking, respectively. Hydrated feed particles may be subjected to a modification of physical structure and to swelling, in this way entrapping additional water. If the assumption is made that the SG of water entrapped is very close to 1.0 (24), the entrapment of water into a swollen particle does not modify the volume of total liquid displaced by the insoluble, saturated OM. Therefore, this amount of water cannot be measured by the pycnometer. The WHC of insoluble OM (grams of H20/g of OM; assumed to be an estimate of the total water retained in the swollen particles) was measured for the washed samples in centrifuge tubes according to the American Association of Cereal Chemists (1) or with a simple filtration technique. With the method of American Association of Cereal Chemists (1), the approximate WHC was preliminarily determined with 5.0 g of material put into preweighed 50-ml centrifuge tubes. After the addition of small amounts of distilled water, the tubes were centrifuged at 2000 x g for 10 min. The supernatant was discarded lightly, and tubes were weighed. Approximate WHC (AWHC) was calculated with the equation: AWHC = [(weight of tube + sediment) - (weight of tube + 5.0)]/5. The WHC then was measured by weighing in four tubes an amount of material as calculated: Journal of Dairy Science Vol. 77. No.3. 1994
776
RAMANZIN ET AL.
Statistical Analyses
15
weight of material = (AWHC + 1) where 15 is the total amount of sample plus water. To the four tubes were added the following volumes of water: 1.5 and .5 ml more and 1.5 and .5 ml less than calculated (15 weight of material). The contents of each tube were mixed for 2 min and then centrifuged as before. The WHC value was calculated as the midpoint between the volumes of water added to the adjoining tubes, respectively, without and with supernatant. For each feedstuff, two WHC determinations were made. With the filtration method, samples were put into duplicate polyester bags and washed and centrifuged, together with tared blank bags, by the procedure described. After washing and centrifuging, blank bags and bags filled with samples were weighed. dried at 60·C for 48 h. and then weighed again. The USG of OM was calculated as the weighted mean of the FSG of insoluble OM (estimated after 0 h of soaking) and the SG (assumed = 1.(0) of water held by the particles (calculated with both methods) by the following equation: USG
+ WHC = FSG 1 + WHC .
The equation assumes that volume changes because of swelling exactly reflect the volume of water being trapped and is analogous to that reponed by Sutherland (25) except that the effect of GV is included in the estimation of FSG.
The FSG data of OM of unwashed and washed feedstuffs were subjected to ANDVA (10) with three factors (feedstuff, washing, and soaking time). The FSG data of soluble OM and GV were subjected to ANDVA with two factors (feedstuffs and soaking time, and feedstuffs and washing, respectively). The WHC data and the estimated USG of insoluble OM were subjected to ANDVA with two factors (feedstuffs and method of WHC determination). Only one factor (feedstuffs) was used for the statistical analysis of WL data. The WL, FSG of soluble and insoluble OM, GV, WHC, and USG data also were subjected to simple correlation analysis (10) with chemical composition of the original feedstuffs. RESULTS WL and FSG
The solubility estimated as WL (Table 2) differed significantly among feedstuffs (P < .(01) and was very low for sugar beet pulp (5.0% OM) and corn gluten meal (12.4% OM). For the other feedstuffs. the WL were higher and ranged between a minimum of 19.9% OM in linseed meal to a maximum of 53.2% DM in corn gluten feed. The FSG of OM of the unwashed feedstuffs and of their insoluble and soluble fractions are also given in Table 2. The FSG of the unwashed feedstuffs and of their insoluble frac-
TABLE I. Chemical composition of the feedstuffs tested. Feedstuff
DM
CP
Fat
Fish meal Soybean meal Flaked soybeans Cottonseeds Linseed meal Com gluten meal Com gluten feed Com meal Barley meal Wheal bran Dehydrated alfalfa meal Dried sugar beet pulp
90.7 88.8 91.1 90.2 91.4 90.2 92.8 87.1 87.9 87.4 91.0 88.3
67.6 52.1 36.3 20.5 32.3 62.9 26.1 10.0 12.9 18.8 15.9 99
8.6 1.6 22.3 7.4 5.5 4.4 3.9 4.0 2.6 1.0 2.7 4.0
NDF
ADF
Cellulose
Lignin
Ash
13.2 10.6 28.2 15.7 3.1 8.8 3.2 7.1 8.8 29.4 28.3
.6 .1 10.3 9.3 .1 1.0 .2 1.2 3.4 8.1 2.\
17.5 7.0 5.4 7.8 8.5 1.7 6.1 1.8 4.5 6.2 10.1 5.2
(% DM)
Journal of Dairy Science Vol. 77. No.3. 1994
19.2 16.4 50.2 394 3.2 35.8 12.7 30.0 44.0 50.0 603
13.8 10.7 38.5 27.5 3.2 11.2 3.4 8.7 12.2 38.4 307
777
WATER-HOLDING CAPACITY AND SPECIFIC GRAVITY
tions ranged from approximately 1.2 to 1.7. The main effect of feedstuff was highly significant. In general, differences in FSG caused by soaking were small, even if the main effect of soaking time was highly significant. The main effect of washing was not significant. All interactions of feedstuffs with washing and soaking time were highly significant. indicating that changes in FSG that were due to washing, soaking time, or both were different for the various feedstuffs. The FSG of the soluble fractions, which were calculated by difference, varied significantly among feedstuffs (P < .(01) from .78 to 2.40 (averages of 0 and 15 h of soaking). The main effect of soaking time was not significant. Also, in some feedstuffs, the FSG of the soluble fractions was largely different from that of the insoluble fractions. GV, WHC, end USG
The estimated GV of the unwashed feedstuffs and of their insoluble fractions are given
in Table 3. The GV were very low and less than 1% of DM in all the feedstuffs, both unwashed or washed; in a few samples, due to variability of measurements, GV results were even negative. The WHC also is listed in Table 3. The main effect of method was highly significant (P < .(01). Centrifuging in centrifuge tubes gave higher values than filtration in polyester bags for all feedstuffs except soybean meal, com meal, and barley meal (interaction of feedstuffs and methods: P < .001). The main effect of feedstuff was also highly significant. However, the differences in WHC among feedstuffs were much higher than the average difference between methods. The WHC ranged approximately from I to 1.5 g of H20/g of DM in com meal and corn gluten meal to approximately 6 g of H20/g of DM in sugar beet pulp. For the other feedstuffs, WHC varied from approximately 2 to 4 g of H20/g of DM. The feedstuffs were ranked in a different order by the methods; the interaction of feedstuffs and method was highly significant (P < .(01).
TABLE 2. Solubility estimated as washing losses (WL) from polyester bags and functional specific gravity (PSG) of DM of the original unwashed feedstuffs and of their insoluble and soluble fractions after soaking for 0 or IS h. FSG of DM Original feedstuff (unwashed) Feedstuff
WL
Insoluble fraction (washed)
Soluble fraction (by difference)
oh
IS h
oh
IS h
oh
IS h
1.48 1.48 1.31 150 1.47 126 1.56 1.62 152 1.23 1.56 1.64
1.52 1.44 1.35 1.60 1.60 1.33 1.53 1.53 1.49 lAO 1.65 1.62
lAS lAS 1.31 151 1.52 1.33 1.48 1.55 1.62 1.44 lAS 1.57
1.50 1.47 1.36 1.60 1.40
1.54 152 1.31 1.48 1.25 .77 1.64 1.75 1.41 .89 1.73 2.95
1.55
(% of DM)
Fish meal Soybean meal Flaked soybeans Cottonseeds Linseed meal Com gluten meal Com gluten feed Com meal Barley meal Wheat bran Dehydrated alfalfa meal Dried sugar beet pulp SE
41.9 35.0 42.8 28.1 19.9 1204 53.2 35.5 47.5 38.2 39.1 5.0 3.9
lAO
1.48 156 1.73 1.60 1.47 1.60
.03
lAO
1.33 1.60 2.41 .79 1.58 1.46 1.23 1.07 1.93 1.84 .35
p
Feedstuffs (F) Washing (W) W x F Soaking time (5) S x F W x S x F
