Supernumeraries in Racing Homer
by Koen Joris (B) and Axel Sell (D)
Speaking about supernumeraries we often get from breeders the
opinion that pigeons with 11 primaries or more have advantages in
the competition against those with 10 primaries (wild-type) only. So
recently in a report in the journal 'Die Brieftaube' no. 44 of 2016.
About the genetics of the phenomenon there was long speculation
only. The perhaps most famous pigeon book in the world, 'The Pigeon'
by Wendell M. Levi, assumes a dominant inheritance. Perhaps because
occasionally some young with this characteristic were raised from
such a parent. For breeders who want to either increase the number
of supernumeraries in their strain or even want to get rid of them,
the results of a recent study will be of interest.

Fig. 1: Supernumerary Racing Homer (Photo: Koen Joris)
Dominant or recessive? In the case of unknown phenomena, this is
often the first question. If it would be a dominant feature we would
get from a heterozygous individual 50% offspring with that feature.
If only 5 or 6 young are raised and none with that feature, then
this could be traced back to chance. If 1 or 2 with that feature are
raised, then one will feel confirmed in his expectation. However,
the same result can also be expected with a recessive feature. If an
individual with one recessive trait is mated with another without
the respective factor in the genome, then you will not get any young
with the trait. If it is accidentally mated with a (recessive)
carrier of the gene, then 50% of the offspring show the trait. A
recessive characteristic for supernumeraries often is assumed from
breeders who occasionally, got an 11-penner in strains where the
other pigeons had only the typical 10 primaries.
Koen Joris was probably the first who did extensive tests. He had
introduced in his loft an 11-penner, which with different hens
raised only young of the wild-type. In the subsequent generations
and after the addition of further 11-penners from other lofts, one-
and two-sided 11-penners were no longer a rarity. From 24 couples of
supernumeraries x wild-type 143 young were raised and 35 or 24% from
them were supernumeraries. For a dominant feature, it should have
been about 50%. With this large number, the deviation can no longer
be traced back to coincidence. The thesis of dominance at Levi could
thus be regarded as disproved.
|
Couples |
Youngsters |
Supernumeraries |
per cent |
Supernumerary cock x Wild-Type hen |
17 |
109 |
27 |
25 |
Wild-Type cock x Supernumerary hen |
7 |
34 |
8 |
24 |
Total |
24 |
143 |
35 |
24 |
Table
1: Supernumerary x Wild-Type und vice versa
It is not yet a refutation for the thesis of a recessive feature.
Because one could not assume that all the phenotypical wild-type
used as a test partner were carrier of the recessive trait. Less
than 50% were therefore expected with a recessive feature. The
thesis, however, could be excluded by the following test series.
First 13 couples of supernumeraries mated with each other produced
71 young and 22 (or 31%) of them phenotypical the wild-type with 10
primaries. In a simple recessive inheritance it would have been
none.
|
couples |
youngsters |
Wild-Type |
per cent |
Supernumerary x Supernumerary Groupe 1 |
7 |
34 |
0 |
0 |
Supernumerary x Supernumerary Groupe 2 |
6 |
37 |
22 |
59 |
Total |
13 |
71 |
22 |
31 |
Table 2: Supernumerary x Supernumerary
The test was supplemented by mating some of the young produced from
supernumerary couples with each other or with their supernumerary
parents (Table 3).
|
couples |
youngsters |
Wild-Type |
per cent |
Supernumerary raised from supernumeraries mated with each
other |
3 |
22 |
3 |
14 |
Supernumerary cock raised from supernumeraries x
supernumerary hen |
2 |
14 |
8 |
57 |
Supernumerary cock x supernumerary raised from
supernumeraries |
4 |
32 |
6 |
19 |
Total |
9 |
68 |
17 |
25 |
Table
3: Supernumeraries with different background mated with each other
At the first test summarized in Table 2 it is striking that 7 of the
couples raised only supernumeraries, while 6 had a significantly
higher proportion of normal ones than those 24% of all 13 couples
together. This indicated a different genetic endowment of the
sub-groups. Some of the supernumeraries appeared to be only
heterozygous with one or several of the factors involved. It was
therefore necessary to look for explanations beyond the simple text
book Mendelian model of heredity. Thus, e.g. in human genetics, a
better understanding of more complex phenomena is attained by
modeling the interaction of several genes. Certain genes can only
have an effect if certain genes are present on one or more other
gene-loci. If they are not present, the trait remains covered. One
speaks of epistatic relationships (Cordell, 2002).
Following the investigation in the strain of racing homers, test
mating were carried out in single boxes by mating some of the
supernumeraries with Gimpel Pigeons. Those were from a stock in
which supernumeraries were not observed so far. Following the idea
of epistatic relations the best statistical fit was derived for a
model of the interaction of three factors sn1, sn2 and sn3. In
combination it seems to be sufficient if one of them is present in
heterozygosity only. Thus, the first generation (F1)
consisted of young with 10 primaries. 130 young of the F2
were raised from three couples of the first generation mated
together over several years. From the F2 only 8 (6.2%)
were supernumeraries.
couples |
youngsters |
supernumeraries |
per cent |
no.
1 |
48 |
2 |
4,2% |
no.
2 |
48 |
3 |
6,3% |
no.
3 |
34 |
3 |
8,8% |
total |
130 |
8 |
6,2% |
Table 4: F2
from gimpel-homer crosses
Finally, 5 of the F1 were mated back to supernumeraries
of the homer stock and produced 12 (20%) from 60 offspring. From all
5 couples at least one supernumerary was raised. The percentages,
however, were different.
couples |
youngsters |
supernumeraries |
per cent |
no. 1 |
27 |
2 |
7,4 |
no. 2 |
8 |
2 |
25,0 |
no. 3 |
10 |
1 |
10,0 |
no. 4 |
8 |
6 |
75,0 |
Nr. 5 |
7 |
1 |
14,3 |
total |
60 |
12 |
20,0 |
Table
5: Backcross of the F1 from supernumerary homer x gimpel
pigeon cross to supernumerary homers

Fig. 2: Backcross supernumerary from Racing Homer x
Gimpel Pigeon cross to supernumerary Racing Homer (Photo: Koen Joris)
More insight will perhaps be gained in other ways with the advances
of molecular genetics. It is hard to imagine, though it would be
desirable that someone will increase the observation base with
conventional pairings. For practical breeding the results mean that
it is easily possible by selection to increase the number of
supernumeraries in a strain and also to breed them true. However,
the introduction of a pigeon with 10 primaries perhaps to stabilize
performance of the strain would be a serious setback for this goal.
On the other hand, breeders, who want to permanently eliminate this
trait from their stock, should not use any supernumeraire in
breeding. The direct offspring will not have the characteristic, but
it will reappear in subsequent generations.
Literature:
Cordell, Heather J., Epistasis: what it means, what it doesn't mean,
and statistical methods to detect it in humans, Human Molecular
Genetics, 2002, Vol. 11, No. 20, pp. 2463-2468.
Joris,
Koen and Axel Sell, Überschwingigkeit bei Haustauben,
RÖK Freude mit
Kleintierzucht, January 2017, pp. 8-11.
Levi,
Wendell M., The Pigeon, 1. ed. 1941, reprinted with changes and
additions 1963, reprinted 1969, Levi Publishing Company, Sumter.
Sell, Axel, Brieftauben und ihre Verwandten, Achim 2014.
Sell, Axel,
Pigeon Genetics, Applied Genetics in the Domestic Pigeon, Achim
2012.
|