Genetic selection increases parthenogenesis in Chinese
painted quail (Coturnix chinensis)1
H. M. Parker,* A. S. Kiess,* J. B. Wells,* K. M. Young,* D. Rowe,† and C. D. McDaniel*2
*Poultry Science Department, and †Experimental Statistics, Mississippi State University, Mississippi State 39762
ABSTRACT Parthenogenesis, embryonic development of an unfertilized egg, occurs naturally in turkey,
chicken, and quail species. In fact, parthenogenesis in
turkeys and chickens can be increased by genetic selection. However, it is unknown if genetic selection for
parthenogenesis is effective in quail or if selection for
parthenogenesis affects egg production. Therefore, the
objectives of this study were to determine if the incidence of parthenogenesis in quail could be increased
by genetic selection and if selection for this trait affects egg production. To prevent fertilization, 1,090 females were caged separately from males at 4 wk of age
and then caged individually at 6 wk of age to monitor
egg production. Eggs were collected daily, labeled, and
stored for 0 to 3 d. After 10 d of incubation, 20 unfertilized eggs from each hen were examined for the occurrence of parthenogenesis and embryonic growth. In
the parent (P) generation and subsequent generations
(1 to 4), hens laying eggs containing parthenogenetic
development and males whose sisters or mothers exhibited parthenogenesis were used for breeding. There was
a linear increase in the percentage of hens exhibiting
parthenogenesis as generation of selection increased.
With each successive generation, there was a quadratic
response in the percentage of eggs positive for parthenogenesis. When compared with the P generation, parthenogenesis was almost 3 times greater for eggs laid
by the fourth generation (4.6 to 12.5%, respectively).
Even when only hens exhibiting parthenogenesis were
examined, the percentage of eggs demonstrating embryonic development responded quadratically with
generation of selection. The embryonic size at 10 d of
incubation was greater for each subsequent generation
when compared with the P generation. There was a
linear decrease in both egg production and the average
position of an egg in a clutch as generation of selection
increased. In conclusion, genetic selection for parthenogenesis increased the incidence of parthenogenesis
and embryonic size but decreased egg production and
average position of an egg in a clutch as generations of
selection increased.
Key words: parthenogenesis, quail, egg production, germinal disc
2010 Poultry Science 89:1468–1472
doi:10.3382/ps.2009-00388
INTRODUCTION
It has been reported that embryonic development in
unfertilized bird eggs, parthenogenesis, occurs in turkeys and chickens (Olsen, 1975), the zebra finch (Schut
et al., 2008), as well as quail (Parker and McDaniel,
2009). However, the majority of the research for parthenogenesis focuses on the Beltsville Small White (BSW)
turkey. For example, Olsen and Marsden (1954) discovered that approximately 14% of the eggs laid by BSW
turkeys exhibited parthenogenesis. However, through
genetic selection using BSW hens exhibiting this trait,
©2010 Poultry Science Association Inc.
Received August 7, 2009.
Accepted April 11, 2010.
1 This is journal article no. J11639 from the Mississippi Agriculture
and Forestry Experiment Station supported by MIS-329160.
2 Corresponding author: [email protected]
the incidence of parthenogenesis was increased with
generation of selection (Olsen, 1975).
To intensify the parthenogenetic trait in BSW turkeys, a selective breeding program was initiated in 1954
(Olsen, 1965b). Olsen selected virgin females that had a
high average incidence of parthenogenesis for his breeding program. At sexual maturity, these females were
mated to males from female lines that exhibited the
highest average incidence of parthenogenesis. By using these selection criteria, Olsen (1965b) increased the
incidence of parthenogenesis in eggs from 14%, before
genetic selection in 1954, to 47.8% after 17 yr of genetic
selection for parthenogenesis (Olsen, 1975).
Parthenogenesis has recently been reported in eggs
from Chinese painted quail (Parker and McDaniel,
2009). Approximately 4.8% of the eggs laid by these
quail exhibited parthenogenesis, and 27% of the hens
used in that study laid at least 1 egg that contained a
form of abortive parthenogenesis. Because these quail
1468
GENETIC SELECTION INCREASES PARTHENOGENESIS IN QUAIL
exhibited the parthenogenesis trait, it is possible that
through genetic selection, the incidence of parthenogenesis can also be increased in these birds.
Parker and McDaniel (2009) also revealed that the
incidence of parthenogenesis was negatively correlated
with egg production in quail. Because egg production
was negatively correlated with parthenogenesis in that
study, it is possible that genetic selection for the parthenogenesis trait could affect egg production. Therefore,
the objectives of this study were to determine if genetic
selection will increase the incidence of parthenogenesis
in the Chinese Painted quail and if genetic selection for
this trait will adversely affect egg production.
MATERIALS AND METHODS
Housing and Care
Quail chicks obtained from parthenogenetic breeding
stock were selected for the subsequent production of
unfertilized eggs for determination of parthenogenesis.
Chicks were fed a commercial quail starter diet until 4
wk of age and were then placed on a commercial quail
breeder diet. Quail were fed ad libitum and exposed
to 17 h of light. Hens (n = 1,076) were separated from
males at 4 wk of age when adult male plumage first became visible. Each hen was individually caged at 6 wk
of age so that initial egg production could be obtained.
All birds were treated in accordance with the Guide
for the Care and Use of Laboratory Animals (NRC,
1996).
Determination of Embryonic Development
and Clutch Position
Daily, individual eggs were collected, labeled, and
stored for 0 to 3 d at 20°C. Eggs were incubated at
standard conditions (37.5°C) for 10 d. The first 20 unfertilized eggs laid by each hen over a 120-d period were
examined at 10 d of incubation for parthenogenesis (Olsen, 1965a; Parker and McDaniel, 2009) using a magnifying lamp at 2× magnification (Philmore, Rockford,
IL). Due to the lack of typical embryonic differentiation
in the majority of the parthenogenetic quail embryos,
embryos examined in this study were not staged using
the procedure of Hamburger and Hamilton (1951). To
determine the degree of embryonic development, under
the magnifying lamp, the blastodisc was measured to
the nearest millimeter across its greatest width (Olsen,
1965a; Parker and McDaniel, 2009). Also, hen-day egg
production was calculated for the first 120 d.
In this study, the first 20 eggs from all hens were examined for the incidence of parthenogenesis and therefore identified hens exhibiting parthenogenesis. The incidence of parthenogenesis in all laying hens examined
was calculated as the percentage of hens previously recorded to have laid parthenogenetic eggs out of the total laying population. The incidence of parthenogenesis
in eggs was determined for all hens examined as well as
1469
for only hens that were previously shown to exhibit the
parthenogenesis trait, as evidenced by the production
of at least 1 parthenogenetic egg.
Hens laying eggs that exhibited parthenogenesis were
used for genetic breeding stock if at least 10% of their
eggs were parthenogenetic from the 20 eggs initially
examined. For the original parent (P) generation, hens
exhibiting parthenogenesis were mated to males from
a randombred population. It was unknown if the dams
of these males exhibited parthenogenesis. However, for
subsequent generations, the males selected for breeding
were from hens that exhibited parthenogenesis in at
least 10% of their eggs. For each generation, unfertilized eggs from the offspring were examined for parthenogenesis using the method described previously. This
method of selection was repeated for each generation.
Statistical Analyses
Data were analyzed as a completely randomized
design with hen as the experimental unit, and means
for each generation of selection were separated using
Fisher’s protected least significant difference (P <
0.05). Regression analyses were used to determine the
relationships of generation of selection with hens positive for the parthenogenesis trait, eggs positive for parthenogenetic development from all hens, eggs positive
for parthenogenesis from hens exhibiting the trait, egg
production, and average position of an egg in a clutch
(Steel and Torrie, 1980).
RESULTS
The percentage of hens laying eggs positive for parthenogenetic development following selection for the
incidence of parthenogenesis in eggs is presented in
Figure 1. When compared with P, generations 2, 3, and
4 contained more hens exhibiting parthenogenesis yet
there was no difference between the P and generation
1. The percentage of hens positive for parthenogenesis
was greater for generation 4 when compared with either
generation 1 or 2 yet was similar to generation 3. There
was a linear increase in the percentage of hens laying
eggs positive for parthenogenetic development as generation increased.
However, there was a quadratic response in the percentage of eggs containing parthenogenetic development as generation of selection increased (Figure 2).
When compared with P as well as generations 1, 2,
and 3, more eggs containing parthenogenetic development were produced by generation 4. Generation 3 had
a higher percentage of eggs laid that were positive for
parthenogenetic development when compared with generations P and 1.
When examining parthenogenesis in eggs that were
laid only by hens that exhibited the parthenogenesis
trait, there was a quadratic response in the percentage of parthenogenetic eggs as generation of selection
increased (Figure 3). Generation 4 laid more eggs ex-
1470
Parker et al.
Figure 1. The percentage of hens laying eggs positive for parthenogenetic development by generation of selection for the incidence of
parthenogenesis in eggs. Each bar represents the percentage of hens
positive for parthenogenesis (n = 1,076; SEM = 4.3). A–DMeans with
different letters are significantly different at P < 0.0001. There was
a linear increase in the percentage of hens exhibiting parthenogenesis
as generation of selection increased (y = 8.00x + 38, r2 = 0.98; P <
0.0016). P = parent generation.
Figure 3. The percentage of eggs that contained parthenogenesis
by generation of selection, when only eggs from hens that exhibited
the parthenogenesis trait were examined. Each bar represents the percentage of eggs positive for parthenogenesis (n = 540; SEM = 1.69).
A,BMeans with different letters are significantly different at P < 0.09.
There was a quadratic response in the percentage of eggs laid exhibiting parthenogenesis as generation of selection increased (y = 0.91x2
− 2.22x + 12.74, r2 = 0.97; P < 0.03). P = parent generation.
hibiting parthenogenesis than generations P, 1, and 2.
However, generation 3 was intermediate when compared with generations P, 1, 2, and 4. When compared
with the P generation, the incidence of parthenogenesis
in eggs produced by hens that exhibited the trait was
not different for generations 1, 2, and 3.
The germinal discs were larger for generation 2 when
compared with generations P, 1, and 3 but were similar
to generation 4 (Figure 4). When compared with the P
generation, all subsequent generations exhibited larger
germinal discs.
The percentage of eggs produced over generation of
selection for parthenogenesis is presented in Figure 5.
There was a linear decline in egg production as bird
generation increased. The P generation produced more
eggs than generations 2, 3, and 4 yet was similar to generation 1. However, generation 1 produced more eggs
than generations 3 and 4.
There was also a linear decrease in the average position of an egg in a clutch as generation of selection for
parthenogenesis increased (Figure 6). The average position of an egg in a clutch was greater for generation 1
when compared with generations 2, 3, and 4 yet similar
to the P generation. Generations 3 and 4 produced less
eggs in a clutch than generations P, 1, and 2.
Figure 2. The percentage of eggs that exhibited parthenogenesis
by generation of selection, when eggs from all hens were examined.
Each bar represents the percentage of eggs positive for parthenogenesis
(n = 1,076; SEM = 1.07). A–CMeans with different letters are significantly different at P < 0.0001. There was a quadratic response in the
percentage of eggs laid exhibiting parthenogenesis as generation of
selection increased (y = 0.47x2 + 0.058x + 4.7, r2 = 0.99; P < 0.01).
P = parent generation.
DISCUSSION
In this trial, for the P generation, the incidence of
parthenogenesis was unknown in males used for mating. However, beginning with generation 2 progeny, the
average incidence of parthenogenesis was known for
both the dams and the dam line of their sires, yet it
was unknown if the males used for mating carried the
parthenogenesis trait. Even though males were initially
chosen from a random population for the P generation,
there was a steady increase in the percentage of hens
exhibiting parthenogenesis as generation of selection increased. In fact, the incidence of parthenogenesis was
almost doubled when comparing hens exhibiting the
trait from the P generation (36.5%) to hens in generation 4 (68.1%). It has been reported that parthenogenesis is a heritable trait in turkeys and chickens (Olsen,
Figure 4. Embryonic development by generation of selection for
the incidence of parthenogenesis in eggs. Each bar represents the diameter of the germinal disc (n = 540; SEM = 0.16). A–CMeans with
different letters are significantly different at P < 0.0001. P = parent
generation.
GENETIC SELECTION INCREASES PARTHENOGENESIS IN QUAIL
Figure 5. The percentage of eggs produced by generation of selection for the incidence of parthenogenesis in eggs. Each bar represents
the percentage of eggs produced (n = 1,076; SEM = 1.5). A–DMeans
with different letters are significantly different at P < 0.0001. There
was a linear decrease in the percentage of eggs produced as generation
of selection increased (y = −2.64x + 30, r2 = 0.99; P < 0.0007). P =
parent generation.
1975), and it now appears to be a heritable trait in the
Chinese painted quail as well.
Interestingly, as generation of selection increased, the
differences in hens exhibiting parthenogenesis between
generations became less. For example, when comparing
the difference between the P generation and generation
1, there was a 10.9 percentage point increase in the percentage of hens exhibiting parthenogenesis, whereas by
generations 3 and 4, this difference between generations
was only 3.7 percentage points. These slight increases
between later generations of selection suggest that the
incidence of parthenogenesis in hens may follow an asymptotic progression, with genetic selection resulting
in even smaller differences between future generations
as selection pressure continues.
These results are similar to what was reported for
BSW turkeys selected for the parthenogenesis trait.
When examining eggs that were positive for parthenogenesis, Olsen (1975) reported a 3-fold increase in
the incidence of parthenogenesis by the fifth generation
of selection. However, Olsen (1975) also revealed that
Figure 6. Clutch size by generation of selection for the incidence
of parthenogenesis in eggs. Each bar represents the average position of
an egg in the clutch (n = 1,076; SEM = 0.26). A–CMeans with different letters are significantly different at P < 0.0001. There was a linear
decrease in the average position of an egg in a clutch as generation
of selection increased (y = −0.61x + 4.14, r2 = 0.83; P < 0.03). P =
parent generation.
1471
the incidence of parthenogenesis in eggs from a random
population versus the first generation was 9 percentage
points, whereas by generation 3 and 4, this difference
between generations decreased to only 4.2 percentage
points. Interestingly, the percentage of eggs positive for
parthenogenetic development in BSW turkeys was never
greater than 46%, even after 12 yr of genetic selection
for parthenogenesis (Olsen, 1975). Possibly, the reason
that the percentage of eggs positive for parthenogenesis
plateaued near 50% in BSW turkey eggs was because
apparently only Z oocytes become parthenogens and
the chance that a hen will ovulate a Z oocyte is 50%
(Olsen, 1975).
In the current study, for the percentage of eggs containing parthenogenetic development, initially as generation increased there was little change. However,
when comparing the percentage of eggs exhibiting parthenogenesis from generation P to generation 3, there
was a 2-fold increase in the number of eggs positive
for parthenogenesis, and by generation 4, this increase
was 3-fold. The results seen in this trial are similar to
what was reported in unfertilized BSW turkey eggs. For
example, after 4 yr of genetic selection for the parthenogenesis trait, Olsen (1965b) increased the incidence
of parthenogenesis in eggs from 14 to 41%. When using Dark Cornish hens, which already exhibited a high
degree of parthenogenesis, Sarvella (1975) reported
that genetic selection for the parthenogenesis trait also
increased the incidence of parthenogenesis. It appears
that genetic selection for parthenogenesis also increases
the number of quail eggs exhibiting parthenogenesis.
Even though the incidence of parthenogenesis was
known for both the dam and sire’s dam beginning with
the progeny in generation 2, the percentage of eggs exhibiting parthenogenesis only increased 1.7% percentage points between generations P and 2 in the current
study. However, the percentage of hens positive for parthenogenesis from generations P to 2 increased over
19 percentage points. In chickens, Olsen et al. (1968)
hypothesized that parthenogenesis is controlled by a
single autosomal recessive gene and perhaps a single
gene may also control whether quail hens exhibit parthenogenesis. Therefore, it is also possible that genetic
selection for a single gene controlling parthenogenesis
would yield rapid increases in the percentage of hens
exhibiting the trait. On the other hand, it may be that
several genes control the actual amount of parthenogenesis exhibited in the eggs of individual hens, possibly resulting in much slower progress in increasing
the incidence of parthenogenesis in the egg by genetic
selection.
Because the incidence of parthenogenesis increased
in both the number of hens that exhibited the trait
and the number of eggs containing development with
genetic selection, it is apparent that parthenogenesis
is a heritable trait in quail. However, it appears that
genetic selection for parthenogenesis also increases the
frequency of parthenogenetic eggs that are laid by hens
that exhibit the trait. For example, when only hens ex-
1472
Parker et al.
hibiting parthenogenesis were examined, 12.5% of the
eggs laid were positive for parthenogenesis in the P
generation stock yet by generation 4, 18.4% of these
eggs exhibited positive development.
The appearance of parthenogenetic embryos from all
generations in the current study was very similar to the
micrographs shown by Parker and McDaniel (2009), in
which they reported an average parthenogenetic disc
size of 3.7 mm in Chinese painted quail. Similarly, in
the present study, the average parthenogen size for the
P generation was 3.3 mm; however, the generation 2
progeny exhibited larger parthenogens than either generations P or 1. In fact, when compared with the P
generation, the germinal disc size was 65% greater for
generation 2. Possibly, because by generation 2 the incidence of parthenogenesis was known for both the dam
and dam line of the sire, blastodiscs from generation
2 exhibited larger parthenogens when compared with
generations P and 1. This increase in parthenogenesis
as generation increased was also noted in BSW turkey
eggs. For example, Olsen (1975) reported an increase
in blood and extraembryonic membranes as well as embryos as he selected hens for the parthenogenesis trait.
Sarvella (1975) also reported an increase in degree of
parthenogenesis in eggs laid by Dark Cornish hens when
selecting for parthenogenesis.
Even though genetic selection increased the percentage of parthenogenetic eggs laid in each subsequent
generation, selection for parthenogenesis reduced the
number of eggs laid by each generation. When compared with the P generation, the percentage of eggs
exhibiting parthenogenesis nearly tripled by generation
4 in this trial; however, generation 4 produced only 66%
as many eggs as did the P generation. Interestingly,
Parker and McDaniel (2009) reported a negative correlation for the percentage of eggs exhibiting parthenogenetic development with egg production.
Parker and McDaniel (2009) also found that the incidence of parthenogenesis was negatively correlated with
clutch size. In the present trial, because the incidence
of parthenogenesis in eggs increased yet egg production decreased with generation, clutch size could have
affected both the percentage of eggs exhibiting development and egg production. For example, when compared with P, generations 3 and 4 exhibited more eggs
positive for parthenogenetic development, yet they had
the shortest average clutch size. Parker and McDaniel
(2009) reported that the incidence of parthenogenesis
was greatest for the first egg in the clutch sequence, and
it is likely that the decrease in egg production in this
study was due to shorter clutch sequences. Therefore,
in the current study, selection for parthenogenesis may
have inadvertently resulted in selection for decreased
clutch length.
In this study, it is evident that selecting Chinese
painted quail on the basis of parthenogenetic eggs affects embryonic development, egg production, and average clutch position. Because these variables are affected
by selection for parthenogenetic eggs, it is possible that
an inadvertent selection for parthenogenesis may negatively interfere with the mechanisms of normal fertilization. As a result, hens that exhibit parthenogenesis
may have poor fertility and higher incidences of early
embryonic death because eggs that exhibit parthenogenesis are similar in appearance to fertile eggs that
exhibit early embryonic death. Also, a reduction in the
number of eggs a hen produces would result in economic
losses. Therefore, a better understanding of parthenogenesis could lead to economic gains in egg production
and possibly fertility.
In conclusion, genetic selection increased the number
of hens that exhibited the parthenogenesis trait, the
number of eggs that exhibited parthenogenesis, and the
size of the parthenogen. Genetic selection also increased
the frequency of parthenogenetic eggs from hens that
were previously shown to exhibit the parthenogenesis
trait. However, selection for the parthenogenesis trait
decreased egg production and average clutch position.
In the current study, because parthenogen embryo size
increased with genetic selection, additional selection for
parthenogenesis in quail may likely result in hatched
parthenogens as found in the BSW turkeys (Olsen,
1975).
REFERENCES
Hamburger, V., and H. L. Hamilton. 1951. A series of normal stages
in development of the chick embryo. J. Morphol. 88:49–92.
NRC. 1996. Guide for the Care and Use of Laboratory Animals.
National Academy Press, Washington, DC.
Olsen, M. W. 1965a. Delayed development and atypical cellular organization in blastodiscs of unfertilized turkey eggs. Dev. Biol.
12:1–14.
Olsen, M. W. 1965b. Twelve year summary of selection for parthenogenesis in Beltsville small white turkeys. Br. Poult. Sci. 6:1–6.
Olsen, M. W. 1975. Avian parthenogenesis. USDA, ARS-NE-65.
USDA, Beltsville, MD.
Olsen, M. W., and S. J. Marsden. 1954. Development in unfertilized
turkey eggs. J. Exp. Zool. 126:337–346.
Olsen, M. W., S. P. Wilson, and H. L. Marks. 1968. Genetic control
of parthenogenesis in chickens. J. Hered. 59:41–42.
Parker, H. M., and C. D. McDaniel. 2009. Parthenogenesis in unfertilized eggs of Coturnix chinensis, the Chinese painted quail,
and the effect of egg clutch position on embryonic development.
Poult. Sci. 88:784–790.
Sarvella, P. 1975. Multiple-yolked eggs from a parthenogenetic stock
of chickens. Poult. Sci. 54:1467–1471.
Schut, E., N. Hemings, and T. R. Birkhead. 2008. Parthenogenesis
in a passerine bird, the Zebra Finch Taeniopygia guttata. Ibis
150:197–199.
Steel, R. G. D., and J. H. Torrie. 1980. Principles and Procedures
of Statistics: A Biometrical Approach. McGraw-Hill Inc., New
York, NY.