AU individual measurements of C02 and CH4 production are tabulated in the appendix. 02 consumption was measured, but caused by temporary technical difficulties in series F in obtaining the same accuracy in the O2 determinations as for CO2 and CH4 the values are not tabulated.
4.1. CO2 and CH4 production in series F, G and H
Mean values of CO2 and CH4 production for each period in series F, G and H are shown in Table 14, 15 and 16, respectively, together with mean values of live weight and intake of organic matter being the pronounced determinants for the CO2 production.
Table 14. CO2- and CH4-production in relation to live weight and intake of organic matter.
Series F. Concentrates + clover-grass hay
Tabel 14. CO2- og CH ^produktion i relation til legemsvægt og optagelse af organisk stof.
Serie F. Kraftfoderblanding + kløver-græs hø
Period no.
I II III IV V VII VIII
Level of
H H H H H L L
n
7 7 8 8 7 6 7
L i i/e weight (kg) Mean
173 189 207 224 242 257 261
SE
2.6 2.8 2.6 2.5 2.8 3.7 3.7
Organic matter
kg
3.21 3.59 3.86 4.26 4.65 2.52 2.56
prod Mean 1611 1821 1957 2125 2350 1552 1547
co2
i. (litres) SE
21.5 32.9 34.3 23.7 28.0 21.2 20.8
prod Mean
108 101 104 114 126 138 147
CH4
1. (litres) SE
14.0 11.3 15.9 15.6 24.2 2.4 3.4
For CO2 production the coefficient of variation (CV = SD/mean value) varied from 2.6% to 5.0% for all series indicating a relatively small variation between calves independent of whether the calves were fed on high or low feeding levels. For CH4 production a much higher variation between calves was found. On high feeding level CV varied from 11% to 51% being reduced to 4-10% for calves on low feeding level.
Table 15. CO2- and CH4-production in relation to live weight and intake of organic matter.
Series G. Concentrates + dried sugar beet pulp + straw
Tabel 15. CO2- og CH^produktion i relation til legemsvægt og optagelse af organisk stof.
Serie G. Kraftfoaerblanding + Rosetter + halm
Period weight (kg) Mean 1. (litres)
SE
. (litres) SE
The CO2 production at 150 kg live weight was about 1500 litres for all series on high feeding level compared with 1000 litres on low feeding level. At 250 kg live weight the CO2 production on high level varied from 2000 to 2400 litres in relation to the intake of organic matter. On low feeding level the CO2 produc-tion was about 1500 litres at a live weight of 250 kg.
Table 16. CO2- and CH4-production in relation to live weight and intake of organic matter.
Series H. Concentrates + clover-grass pellets + straw
Tabel 16. CO2- og CH ^produktion i relation til legemsvægt og optagelse af organisk stof.
Serie H. Kraftfoderblanding + kløver-græs piller + halm
Period weight (kg) Mean . (litres)
SE 1. (litres)
SE
CH4 production ranged from 50 to 150 litres dependent on live weight, feeding level and source of reoughages. The CH4 production in relation to CO2
production was 5.7 and 9.2% in series F on high or low feeding levels,
respect-ively. In series G, the corresponding values were 4.6 and 8.0%, while no difference between levels was found in series H where the CH4 production was about 7.6% of the CO2 production.
4.2. Prediction of CO2 production in growing calves
In the present investigation a linear relationship between CO2 production and metabolic live weight was indicated for the different series. By using all individ-ual measurements, regressions of CO2 production on metabolic live weight have been calculated for all calves on the high feeding level (n — 97) and on the low level (n = 76) with the following results:
(1) High level: CO2, litres = -237 + 37.9 W,kg0 7 5
sT and sb = 116 2.19
RSD = ±178 (CV = 10.2%) (r2 - 0.759) (n= 97) (2) Low level: CO2, litres - -300 + 28.6 W,kg075
sT and sb = 8 1 1.46
RSD = ±131 (CV = 10.4%) (r2 = 0.838) (n = 76) The difference between the two equations was highly significant (P<0.001), but with high RSD for both equations (CV about 10%), the use of the equations is problematic.
The CO2 production is strongly influenced by the intake of organic matter and for that reason a new set of regressions of this relationship was calculated as:
(3) High level: CO2, litres = -180 + 0.54 IOM, g
SÏ and sb = 46 0.013
RSD = ±81 (CV = 4.6%) (r2 = 0.950) (n = 97) (4) Low level: CO2, litres = -7.8 + 0.58 IOM, g
SÏ and sb - 37.6 0.017
RSD = ±78 (CV = 6.2%) (r2 - 0.942) (n = 76) Compared with the first set of equations (1) and (2), the values for r2 have increased considerably and the standard deviation of residual (RSD) is now 4.6% and 6.2% on high or low level, respectively, being acceptable. The difference between the two equations was highly significant (P<0.001).
Finally a regression on both metabolic live weight and intake of organic matter (IOM) have been calculated with the following résulta:
(5) High level: CO2, litres - -202 + 1.82 W,kg075 + 0.52 IOM,g
S!andsb = 53 2.14 0.027
RSD = ±81 (CV = 4.6%) (r2 = 0.951) (n - 97)
(6) Low level: CO2, litres - -191 + 9.78 W,kg0-75 + 0.42 IOM,g SiandSb = 37 1.27 0.024 RSD = ±58 (CV = 4.6%) (r2 = 0.968) (n = 76)
By including metabolic live weight and organic matter in the regressions the equation for calves on the high feeding level was not improved but the equation for calves on low level, being near maintenance level, was slightly improved.
The standard deviation of residuals decreased to 4.6% and r2 increased to 0.968.
4.3. Discussion
Comparing the results obtained in measuring the CO2 production and CH4 production it is obvious that the variation between animals concerning their CH4 production is much greater than for CO2 production, confirming the results of Schiemann, Jentsch, Wittenburg and Hoffmann (1976). The CO2
production depends mainly on the maintenance metabolism of the animals and their intake of feed, while the CH4 production is strongly influenced by the fermentative processes in the rumen depending on the type of concentrates and roughages applied. This indicates that a greater variation for CH4 production than for CO2 production as found must be expected.
No significant differences in CO2 production in relation to metabolic live weight were found between series F, G and H compared on the same levels of feed intake. By pooling the individual measurements and by regression of CO2
production on metabolic live weight two equations (1) and (2) were obtained for the two feeding levels, but with CV-values of about 10% the use of the equa-tions is problematic.
Regression of CO2 production on intake of organic matter (IOM) improved the equation considerably (3) and (4). With CV = 4.6% and 6.2% on high or low level, respectively, the accuracy was now comparable to the one obtained in experiment with pigs (Thorbek, 1975). By including both metabolic live weight and intake of organic matter the equation on low feeding level (6) was further improved (CV = 4.6%). On high level no improvement was found and the regression coefficient for metabolic live weight was not significant.
The equations found indicate that for calves on high feeding level (near ad lib.
feeding) the intake of organic matter can be used to predict the CO2 production while on low feeding level (near maintenance level) metabolic live weight should be included in the equation.
By using all data concerning CH4 production and live weight from calves on the low feeding level, where the coefficients of variation were between 4 and 10%, and plotting CH4 production on live weight a linearity was indicated and the CH4 production was found to be about 70 litres at 150 kg live weight increasing to about 120 litres at 250 kg live weight. A regression gave CH4, litres = 0.49 W, kg but with CV = 18.4% and r2 = 0.746 the use of the equation is problematic.
Caused by the great variation in CH4 production between calves fed on the high feeding level no reliable regression of CH4 production on live weight could be established (r2 = 0.19). Nevertheless, plotting the data in the 3 series against live weight it can be demonstrated that CH4 production was lowest in series G where the roughage consisted of dried sugar beet pulp + straw. The CH4
production at a live weight of about 190 kg was in series G 75 litres compared with a production of 101 and 110 litres in series F and H, respectively, and the differences were highly significant.
4.4. Conclusions
1. Three experiments, each with eight growing bull calves (Holstein-Friesian) were made in which the same concentrate mixture was given on high feeding level or low level near maintenance. Three different sources of roughage were applied: clover-grass hay (Ser. F), dried molassed sugar beet pulp + barley straw (Ser. G) or clover-grass pellets + barley straw (Ser. H). Seven to eight balance periods, each consisting of 7 days' collection of faeces and urine with a 24 hours measurement of the gas exchange in the middle, were performed with each calf.
2. At 150 kg live weight the CO2 production was about 1000 litres on mainten-ance level and 1500 litres on high feeding level. At 250 kg the respective values were 1500 litres and 2200 litres. The coefficients of variation were rather low, with CV-values between 2.6% and 5.0%.
3. The CH4 production varied from 50 to 150 litres with great variation between animals. On high feeding level the coefficients of variation (CV) were between 11 and 51%, being reduced to 4-10% on low feeding level.
4. Om high feeding level the best fitting regression equations for prediction of CO2 production was CO2, litres = -180 + 0.54 IOM, g, based on intake of organic matter (IOM), while on feeding level near maintenance the metab-olic live weight should be included in the equation as: CO2, litres = -191 + 0.42 IOM, g + 9.78 W, kg075.