Maataloustieteellinen A ikakauskirja
Vol. 60: 599—601, 1988

Litter live weight at birth with varying levels of D’Man germplasm

M. BOURFIA 1 and R. W. TOUCHBERRY 2
'Department of Animal Production, I.A.V. Hassan II
P.0.8 6202, Rabat-Instituts, Morocco

department of Animal Science, University of California
Davis, CA 95616, USA

Abstract. A total of 208 litter birth weights, with live lambs produced in the northwest
of Morocco was analyzed both by factorial analysis and by multiple regression analysis using
least squares procedure. In the first model, breed group of dam and age of dam were included
as fixed effects. In the second model, maternal additive and heterotic effects were included
instead of breed group of dam.

Results of the first analysis indicated that breed group of dam influenced litter live weight
at birth (LLWB). The crossbred S x D dams (with the sire breed listed first), and the purebred
D’Man dams produced the heaviest live litters, in average 4.25 and 4.08 kg, respectively. The
purebred Beni Guil dams produced the lightest live litter, 3.32 kg in average. The second anal-
ysis showed that maternal additive effects were important for LLWB, and that maternal hete-
rosis was of little importance. The ranking of maternal additive breed effects for LLWB was
D’Man, Sardi, and Beni Guil in descending order.

Index words: Additive effect, heterosis, sheep crossbreeding, D’Man

Introduction

Effective use of prolific sheep in crossbreed-
ing systems is based on exploitation of breed
and heterotic effects. Little effort has been ex-
panded in quantifying additive and heterotic
effects of Moroccan breeds of sheep. The ob-
jectives of the study were to evaluate lamb-
ing performance, compare maternal additive
breed differences, and assess maternal hete-
rosis for litter live weight at birth (LLWB)

with different proportions of D’Man (D)
breeding.

Material and methods

D sires were mated to nine breed groups of
ewes resulting from a diallel cross conducted
at a station located in the northwest of Moroc-
co (latitude 34°18 N). The base diallel includ-
ed three Moroccan breeds of sheep, i.e. D (D),
Beni Guil (G), and Sardi (S). The trait stud-

599

JOURNAL OF AGRICULTURAL SCIENCE IN FINLAND



ied was LLWB, which was the sum of birth
weights of lambs born in the same litter iden-
tified as live within 20 hours of birth. A total
of 208 litter birth weights recorded in the
winter of 1985 was analyzed by least squares
procedure (3), using two models. In the first
model, genetic effects were included as breed
group of dam, while in the second model,
genetic effects were defined in terms of mater-
nal additive and heterotic effects. In the lat-
ter model, fractions based on the expected
proportions of genes contributed by each
breed for maternal additive and heterotic ef-
fects were computed as shown in table 1, and
were considered as continuous variables; the
genetic components were then estimated as
partial regression coefficients (2, 5,6, 7). Be-
cause of dependencies among breed additive
effects, maternal additive breed effects for G
and S were computed as deviations from D.
Both models included environmental effects
as age of dam with two levels, i.e. 2 and 3 year
old dams.

Results and discussion

LLWB represented a composite trait of

prolificacy, viability, and prenatal growth,
and may be regarded as an early indicator of
ewe productivity. The two statistical models
used in the study can be derived from the ana-
lyses of variance given in table 2. The cor-
responding least squares means and regres-
sions are provided in table 3. Age of dam was
not a significant source of variation for
LLWB.

Dam breed group differences. The first
analysis showed that breed group of dam sig-
nificantly influenced LLWB. Duncan’s mul-
tiple range test as modified by Kramer (4)
revealed 3 groups of mating combinations as
shown in table 3. The crossbred SxD (sire
breed first) and the purebred D dams demon-
strated the highest LLWB with more than
4 kg. The lowest performing group included
the crossbred G x D, G x S, and S x G, and the
purebred G dams. Live litters out of the lat-
ter group of dams were in average more than
.5 kg lighter than live litters from the former
group of dams.

Dam breed group components. The infer-
ences drawn from the results can be extended
after partitioning the breed group effects into
its components. Among the possible reasons

Table I. Coefficients for the fractions of maternal additive and heterotic effects that were used in model 2.

DCS DG GD DS SD GS SG

Mat. Additive D 1 0 0 1/2 1/2 1/2 1/2 0 0
GO 1 0 1/2 1/2 0 0 1/2 1/2
SOO 1 00 1/21/2 1/2 1/2

Mat. Heterosis 0 0 0 111111

D =D’Man, G = Beni Guil, and S = Sardi. Breed of sire is listed first in the case of crossbred dams.

Table 2. Least squares analysis of variance for litter live weight at birth using two statistical models.

Model 1 Model 2

Source D.F. M.S. Source D.F. M.S.

Dam age 1 .005 Dam age 1 .0004
Dam breed group 8 2.252* Dam breed group comp.

Mat. Add. G 1 13.493"
Mat. Add. S 1 1.235
Mat. Heteros. 1 .050

Remainder 198 1.042 Remainder 203 1.033

*:P<.O5 **:P<.ol

600



for differences in performance of dam breed
groups are maternal additive and heterotic ef-
fects. Table 2 shows that maternal additive
breed effects were the predominant genetic in-
fluences. Maternal additive effects for the G
and S, expressed as a deviation from D, were
negative for LLWB, although only the G had
significantly low performance (table 3). The
—.874 kg G maternal additive influence ac-
counts for the generally lower prolificacy of
G ewe. Under the same management system,
Bourfia (1) estimated average litter size at
birth as 2.01, 1.00, and 1.11 for D, G, and
S ewes, respectively. Maternal heterosis was
not statistically significant. It should be point-
ed out that only average maternal heterosis
was estimated, and it is possible that the mag-
nitude of maternal heterosis in certain breed
combinations might be of greater importance.
Since only a relatively few dam breed groups
were available, specific maternal heterosis
could not be estimated because of dependen-
cy in the data.

Conclusion

The multiple regression approach provid-

Table 3. Least squares means, regressions, and standard
errors for litter live weight at birth (kg).

Obs. L.S.M. S.E.

Dam breed group
Sardix D'Man 15 4.252 a .276
Purebred D'Man (D) 41 4.081 a .161
D'Man xSardi 21 3.886 ab .225
D'Man xß.Guil 19 3.880 ab .234
Purebred Sardi (S) 18 3.853 ab .245
B.GuilxD'Man 33 3.509 b .188
B.Guilx Sardi 24 3.453 b .210
Sardi xß.Guil 18 3.444 b .242
Purebred B.Guil (G) 19 3.323 b .237

Dam breed group components
Maternal additive for G (1) —.B74** .242
Maternal additive for S (1) —.261 .239
Maternal heterosis —.033 .152

Obs = Number of observations. Breed of sire is listed first
in the case of crossbred dams. Means followed by
the same letter do not differ (P>.os).

(1) As deviation from that of the D'Man. ** : P<.ol.

ed a useful supplement for interpreting the
differences among the various breed groups
of dams. Maternal additive effects were the
main causes of variation among dam breed
groups for LLWB. On the basis of the mag-
nitude of differences in maternal additive ef-
fects, it seems that use of D germplasm would
increase LLWB.

References

1. Bourfia, M. 1987. Evaluation of three Moroccan
breeds of sheep. PhD. Thesis. University of Minneso-
ta. St Paul. USA.

2. Dillard, E.U., O. Rodriguez and O.W. Robison.
1980. Estimation of additive and nonadditive direct
and maternal genetic effects from crossbreeding beef
cattle. J. Anim. Sci. 50: 653—663.

3. Harvey, W.R, 1977. Users guide for LSML 76.
Mixed model least squares and maximum likelihood
computer program. Ohio State University, Columbus.
USA.

4. Kramer, C.Y. 1976. Extension of multiple range test
to group means with unequal numbers of replications.
Biometrics, 12: 307—310.

5. Olson, T.A., A.V. Dijk, M. Koger, D.D. Har-
grove, and D.E. Franks. 1985. Additive and hete-
rosis effects of preweaning traits, maternal ability and
reproduction from crossing of the Angus and Brown
Swiss breeds in Florida. J. Anim. Sci. 61: 1121 —1131.

6. Roberson, R.L., J.O. Sanders and T.C. Cart
wright. 1986. Direct and maternal genetic effects on
preweaning characters of Brahman, Hereford and
Brahman-Hereford crossbred cattle. J. Anim. Sci. 63:
438—446.

7. Robison, 0.W., B.T. McDaniel and E.J. Rincon.
1981. Estimation of direct and maternal additive and
heterotic effects from crossbreeding experiments in
animals. J. Anim. Sci. 52: 44—50.

601