Introduction
Despite an abundance of waterfowl worldwide (154
species are currently recognized; Ellis-Joseph et al., 1992),
the survival of many individual species is threatened.
According to a Conservation and Assessment Management
Plan Workshop for waterfowl (Ellis-Joseph et al., 1992),
species under threat of extinction include the Hawaiian
goose or Nene (Branta sandvicensis), Hawaiian duck (Anas
wyvlliana), Laysan teal (Anas laysanensis), Tule goose
(Anser albifrons gambelli) and Trumpeter swan (Olor buc-
cinator). Other species, like the Northern pintail (Anas
acuta) are subject to non-cyclical population decreases or
local?regional range contractions, indicative of species or
populations that are not yet endangered, but may become
so in the near future (US Fish and Wildlife Service, 1999). 
As populations decrease in size (in nature or captivity),
inbreeding can occur which, in turn, can reduce reproduc-
tive fitness (Nei et al., 1975), including fecundity and sur-
vivability (Ralls and Ballou, 1986). When genetic variation
is maintained or increased, population vigour and the 
ability to adapt to environmental change are enhanced.
Challenges in waterfowl management include cross-
species hybridization (Ankney et al., 1986) that can result
in a loss of species uniqueness and specialized adaptive
and fitness traits, and failure to reproduce in captivity
owing to behavioural problems associated with confine-
ment (Hediger, 1965). One solution to these management
challenges is the use of assisted breeding techniques in-
cluding sperm preservation combined with artificial in-
semination (AI). Such strategies can complement captive
breeding programs by: (1) preserving valuable genes from
individuals for infusing into future generations; (2) reduc-
ing disease transfer risk by transporting genetic material 
instead of birds between flocks; and (3) providing a reposi-
tory of genetic material for species not yet endangered that
may be threatened eventually by unexpected hybridization
(with common species) or by natural catastrophe.
Although information is available on basic sperm physiol-
ogy from agriculturally valuable birds, especially chickens
and turkeys (Bakst, 1990), there is a dearth of information
on non-domestic birds. If AI is to be an effective tool for
the management and conservation of rare waterfowl
species, a high priority should be basic reproductive stud-
ies that focus on consistent sperm collection and sperm
sensitivity to cooling and freeze?thawing.
Short-term (6 h) preservation of semen has been used
extensively in the turkey industry (Sexton, 1981; Bakst,
1990), but has been shown to result in reduced fertility
over time (Wishart, 1984; Sexton, 1988a; Thurston, 1995).
Several studies have addressed the issue of developing 
cryopreservation techniques for domestic birds, including
domestic fowls (Polge, 1951) and domestic drakes (Tai et
al., 1983; Tselutin et al., 1995; Watanabe et al., 1981).
Characterization of Northern pintail (Anas acuta) ejaculate 
and the effect of sperm preservation on fertility
L. M. Penfold, V. Harnal, W. Lynch, D. Bird, S. R. Derrickson
and D. E. Wildt
Conservation and Research Center, National Zoological Park, Smithsonian Institution, 
Front Royal, VA 22630, USA
Reproduction (2001) 121, 267?275 Research
Northern pintail duck semen and sperm traits were charac-
terized, and the fertility of cold-stored spermatozoa was in-
vestigated using artificial insemination. Excellent quality
ejaculates containing high proportions of motile spermato-
zoa were collected from drakes within 20 s by a massage
technique. Semen was collected in Beltsville poultry semen
extender, pooled and cold-stored (48C) for 0, 24, 48 or
72 h. Hens were inseminated with 100 m l twice a week,
and eggs were assessed for fertilization and hatch success.
Fertilization success was similar (P > 0.05) for semen cold-
stored for 0 (51.6%), 24 (51.5%), 48 (41.1%) and 72 h
(22.3%; P > 0.05). Similar (P > 0.05) percentages of fertil-
ized eggs hatched to live offspring (73.1, 71.4, 87.0 and
80.0%, respectively). Fresh semen was also equilibrated
with 1 or 4% dimethylsulphoxide or glycerol, and cryopre-
served at the following rates: (1) approximately 608C min?1
(in liquid nitrogen [LN2] vapour) for 10 min; (2) 18C min?1
to ?208C, LN2 vapour for 10 min; and (3) 18C min?1 to
?358C, all followed by immersion in LN2. After thawing for
30 s at 378C or 20 min at 48C, sperm motility and viability
were assessed. The highest numbers of motile spermatozoa
were recovered after slow?fast freezing (2) and thawing at
08C (P < 0.05), but survival was inadequate to allow artifi-
cial insemination. Nonetheless, cold storage provides an 
effective means of short-term storage with no loss of 
fertility in this waterfowl species.
? 2001 Journals of Reproduction and Fertility
1470-1626/2001
Email: Lindap@wogilman.com
However, few studies have been carried out on non-
domestic species (Sexton and Gee, 1978; Gee, 1983) and
fewer still for non-domestic waterfowl (Aleutian Canada
goose (Branta canadensis); Gee and Sexton, 1990).
The Northern pintail (Anas acuta), a common North
American dabbling duck, was used in the present study as
a research model for non-domestic waterfowl. This paper
describes semen collection efficiency and characterizes
sperm traits in a typical Northern pintail ejaculate. Semen
handling techniques and the effects of cold storage (short-
term preservation) and cryopreservation (long-term preser-
vation) on duck sperm motility and viability were also
investigated. Finally, the effect of short-term sperm preser-
vation on fertility after AI was examined. 
Materials and Methods
Animals
Bachelor groups of adult Northern pintails (five groups
of 9?10 each, total of 46 birds; 2?6 years, approximately
850 g each) were housed in outside pens (20 m 3 70 m),
each containing a pond. Six groups of four adult, female
conspecifics (n = 26; 2?6 years, approximately 850 g
each) were housed in pens adjacent to males, separated by
a chain link fence allowing all birds visual and vocal cue
exchange. Individuals in each group were fitted with num-
bered leg bands for individual identification. Four breed-
ing pairs of Northern pintails were maintained as natural
mating controls in similar pens and environmental condi-
tions. All birds were provided with Sinking Duck Diet
(Zeigler Bros. Inc., PO Box 95, Gardners, PA) ad libitum,
but this was removed from the pens containing the males
approximately 12 h before semen collection. Project ap-
proval was granted by the Conservation and Research
Centre?s Institutional Animal Care and Use Committee.
Nesting and egg laying observations
Pens contained adequate cover for nest building, in-
cluding long grass, tufts of pampas grass and small bushes.
Females were observed daily for nest building activity, and
the resulting nests were checked daily for eggs. After dis-
covery of the first egg, each consecutive egg (laid daily)
was numbered with a waterproof marker pen. The number
of eggs present in the nest before AI began was noted to
ensure that only eggs produced after AI would be consid-
ered in the fertility assessments. Nests in control (natural
mating) pens were similarly checked, and eggs were num-
bered. Clutches were designated complete when no new
eggs were found after 2 consecutive non-laying days. 
Ejaculate collection and processing
Semen was collected from each bird during April?June,
approximately twice a week, for two seasons. Males were
caught using a long-handled net and retrained in a supine
position. Semen was collected by massaging the ventral
aspect of the abdomen towards the cloaca (Gee, 1983).
Further massage at the cloacal base caused semen to pool
at the cloacal opening. Eversion of the copulatory organ,
which could have caused loss of semen over the surface,
was avoided. In general, an ejaculate was obtained within
20 s of massage onset, and non-responsive birds were re-
leased within 1 min to prevent handling stress. A tom-cat
catheter (Sherwood Medical, St Louis, MO) trimmed to
4 cm in length and attached to a 1 ml tuberculin syringe
was used to aspirate semen from each bird. According to
individual experiments (described below), semen was col-
lected neat or immediately placed into Beltsville poultry
sexton extender (BPSE; Sexton, 1977) with the osmolarity
adjusted to 280 mOsm. All samples were transported
within 30 min to the laboratory at ambient (248C) or cool
(48C) temperature (on ice) after placing the catheter in a
glass tube in a Styrofoam container.
Semen and sperm assessments
Semen and sperm analyses were carried out initially on
mid-season ejaculates from different drakes. Semen volume
was determined by aspiration into a calibrated, positive dis-
placement Eppendorf pipette (Brinkman Instruments Inc.,
Westbury, NY). Indicator strips (ColorpHast, EM Science,
Gibbstown, NJ; pH range, 6.5?10.0) were used to deter-
mine the pH of neat semen, and osmolarity was deter-
mined with a vapour pressure osmometer (Westcor Inc,
Logan, UT). Sperm concentration was assessed by diluting
neat semen 1:800 with 0.3% glutaraldehyde in PBS and
counting the number of spermatozoa in a 10 m l volume
using a haemocytometer. The percentage sperm motility
and progressive motility (on a scale of 0?5; 0 = non-motile,
5 = highly motile; Howard et al., 1991) were assessed sub-
jectively by diluting neat semen 1:100 and extended semen
1:10 with BPSE (final concentration approximately 1:100).
Semen dilution with BPSE enhanced the ability to estimate
motility accurately in the normally densely concentrated
ejaculate (Mann, 1964). Aliquots (5 m l) were placed on a
pre-warmed (388C) microscope slide, covered with a cover
slip (22 3 22 mm) and examined under a Leitz Diaplan mi-
croscope (3 20; Leica, 6330 Wetzlar). The mean motility
of three fields was estimated for each slide. Sperm mor-
phology was assessed by incubating 2 m l of neat or ex-
tended semen (depending on the study) in 20 m l of
eosin?nigrosin solution (Burrows and Quinn, 1939) for 2
min. Smears were made on microscope slides, air-dried,
and 100 spermatozoa were assessed (3 40) for the type
and proportions of abnormal spermatozoa (pleiomor-
phisms). Viability, defined as spermatozoa with an intact
plasma membrane, was assessed by incubating 200 m l
semen diluted 2:1 with BPSE with Live?Dead stain consist-
ing of two components, A and B (Molecular Probes, Inc.,
Eugene, OR; Garner et al., 1994; Kasai et al., 1996). A
working solution of component A was made by adding 5 m l
stock solution (SYBR-14) to 95 m l BPSE (working solution).
Diluted semen (200 m l) was incubated for 10 min with 2 m l
268 L. M. Penfold et al. 
component A working solution and 2 m l stock component
B (propidium iodide). Spermatozoa that stained green were
considered viable, whereas those that stained red or 
partially red were assessed as nonviable (Garner et al.,
1994; Kasai et al., 1996).
Artificial insemination of females
Inseminations began after each female had laid the first
egg. Each hen was monitored daily for onset of egg pro-
duction, which opened the reproductive tract and facili-
tated insertion of the insemination catheter. Each female
was captured using a long-handled net and manually re-
strained. The abdomen was gently palpated to ensure the
absence of a shell-covered egg in the oviduct that could
prevent semen from reaching the next unfertilized (un-
shelled) egg. Care was taken during capture, restraint and
palpation to avoid breaking a hard-shelled egg within the
tract. Extended semen was placed intracloacally using a
turkey insemination straw (IMV International Corp.,
Shingle Creek Parkway, MN) attached to a 1 ml tuberculin
syringe (A. Daigger and Company, Heathrow Drive, IL)
with a short length of plastic tubing (A. Daigger and
Company). Inseminations (100 or 200 m l) containing an
average of 3.7?7.4 3 109 spermatozoa, of which approxi-
mately 90% were motile, see below) were performed
twice a week (every 3?4 days) until the clutch was com-
plete. Hens were allowed to incubate complete clutches
for 5?7 days, and then eggs were removed, candled to de-
termine fertilization (assessed by observing a small, dark
patch surrounded by blood vessels inside the egg) and
transferred to an artificial incubator. Approximately 2?3
days before hatching (total incubation time, 21?22 days),
eggs were transferred to a surrogate hen and allowed to
hatch. Fertilization success was defined as the percentage
of eggs produced after AI commenced that were candled
and seen to be fertilized. Hatch success was determined as
the number of live ducklings produced from the total num-
ber of eggs produced after AI commenced. 
Experiment 1: semen characteristics and sperm motility
after ejaculate collection with and without BPSE at
248C or at 48C
Ejaculates were collected from 20 individual drakes,
transported to the laboratory and assessed for percentage
motility, progressive motility and percentage normal sperm
morphology.
Semen was collected from three groups of eight ran-
domly selected drakes, four times a week over 5 weeks,
and each group was pooled to examine the effects of col-
lection and transport to the laboratory on sperm quality.
Pooled ejaculates were assigned randomly to the following
treatments: ?neat?, extended in 100 m l BPSE at ambient
temperature (248C) or extended in 100 m l BPSE and placed
in a Styrofoam box on ice (48C). After transport to the labo-
ratory (which took approximately 40 min) using the three
approaches, spermatozoa were immediately assessed for
percentage motility, progressive motility and percentage
normal sperm morphology.
Experiment 2: fertility and hatchability of eggs after
artificial insemination with extended semen versus
natural mating
After the results of Expt 1, semen was collected from
randomly selected drakes (n = 6?8) into tom-cat catheters
pre-loaded with 100 m l BPSE at ambient temperature and
immediately transferred to a glass tube on ice in a
Styrofoam container. Individual ejaculates were pooled
and assessed for per cent sperm motility, progressive
motility, sperm concentration and viability. Hens were 
inseminated twice a week with 100 or 200 m l of freshly
collected, cooled (48C), extended semen. Nests were mon-
itored daily for egg production, and fertility and hatchabil-
ity were determined and compared with naturally mated
controls. 
Experiment 3: fertility and hatchability of eggs after
artificial insemination with 100 m l of extended cold-
stored semen
After the results of Expt 2, semen was collected from
randomly selected drakes into tom-cat catheters pre-
loaded with 300 m l BPSE to ensure sufficient extension for
subsequent assessments and inseminations. Individual
ejaculates were pooled (n = 6) and assessed for percentage
sperm motility, progressive motility and viability. Sperm
concentration was also noted to ensure inseminates con-
tained similar numbers of spermatozoa. Pooled semen was
divided into four aliquots, three of which were transferred
to a refrigerator (48C) for 24, 48 or 72 h. The remaining
aliquot was placed on ice, designated as 0 h and used im-
mediately for AI. At 24, 48 and 72 h, the assigned cold-
stored aliquots were used for AI. Six pens of four females
were assigned to treatment groups as follows: one pen for
each of the 24 and 48 h treatments and two pens each for
the 48 and 72 h treatments. Females in each pen were in-
seminated with extended semen from the same treatment
group, which avoided confusion in the event of a female
?parasitizing? the nests of other females (that is, one female
laying eggs in the nest of another). Groups of females were
inseminated twice a week with 100 m l extended semen.
Eggs were incubated, assessed for fertilization and hatch-
ing success as described above. 
Experiment 4: impact of freezing and thawing methods
on sperm motility and viability
Pooled semen was subjected to three different freezing
rates in three separate trials. Over the course of 3 weeks,
ejaculates were collected weekly from eight males into
catheters pre-filled with 100 m l BPSE. After transfer to the
laboratory and pooling, aliquots (200 m l) were transferred
to five microcentrifuge tubes (1.8 ml; Perfector Scientific,
Atascadero, CA). An equal volume of BPSE containing 8%
Preservation of waterfowl spermatozoa 269
dimethylsulphoxide (DMSO), 2% DMSO, 8% glycerol, 2%
glycerol or no cryoprotectant (v/v; control) was added to
yield final concentrations of 4 and 1% DMSO, 4 and 1%
glycerol and 0% cryoprotectant, respectively. The osmo-
larity of the cryoprotectants was 1370 mOsm (4% DMSO),
460 mOsm (1% DMSO), 1480 mOsm (4% glycerol),
630 mOsm (1% glycerol) and 280 mOsm (control).
Extended semen was mixed by inverting the microcen-
trifuge tubes three times before incubating at 48C for a 
further 15 min (Gee and Sexton, 1990). The final concen-
tration of extended semen was adjusted to approximately
1 3 109 sperm ml?1. Each sample was assessed for per-
centage sperm motility, progressive motility and viability
to determine the effect of adding cryoprotectant. After
loading and sealing 0.25 ml French straws (TS Scientific,
Perkasie, PA), spermatozoa were cryopreserved by (1) ?fast
freeze?, whereby straws were transferred to 2.5 cm above
liquid nitrogen (LN2) vapour and frozen at a rate of 60?C
min?1 from 4?C to ?70?C, before plunging into LN2; (2)
?slow?fast freeze?, whereby straws were transferred to an
ice?alcohol freezing unit (Biocool, FTS Systems Inc., Stone
Ridge, NY) and cooled from 48C to ?208C at a rate of 18C
min?1, transferred to 2.5 cm above LN2 and cooled at ap-
proximately 608C min?1 to ?70?C before plunging into LN2
(Gee and Sexton, 1980); and (3) ?slow freeze?, whereby
straws were transferred to an ice?alcohol freezing unit and
cooled at a rate of 18C min?1 from 48C to ?358C before
plunging into LN2 (Lake and Stewart, 1978). Freezing rates
were determined by placing a thermistor probe (Fisher
Scientific, Orlando, FL) inside a 0.25 ml straw and reading
the temperature at 15 s intervals. Straws were stored in
LN2 for a minimum of 24 h before thawing using a ?fast?
method (transferring to a water bath at 378C for 30 s) or a
?slow? method (transferring to an ice water bath at 48C for
20 min). After thawing, straw contents were released into a
1.8 ml centrifuge tube, warmed to room temperature (ap-
proximately 248C) and assessed for percentage sperm
motility, progressive motility and viability. Thawed motile
spermatozoa were maintained overnight at 48C and as-
sessed for motility and status after 24 h.
Statistical analysis
Data are presented as means 6 SEM. Differences between
percentage sperm motility, progressive motility, percentage
viability, percentage normal morphology and sperm con-
centration were examined by ANOVA, after testing the data
for normality, using a multifactorial procedure and Fisher?s
least squares means post hoc test (Systat, 1989). Egg fertil-
ization, embryo loss and hatching data were analysed using
Student?s t test or Mann?Whitney U test.
Results
Ejaculate collection, semen and sperm traits 
During the first season, the efficacy of semen collection
was recorded. A total of 180 ejaculates was collected from
46 drakes, and 84% of these were collected within 20 s.
Semen and sperm traits for the Northern pintail during the
mid-breeding season are shown (Table 1; 20 typical ejacu-
lates from 20 different drakes). Good quality semen was typ-
ically cream or white in appearance. In a parallel study that
examined seminal traits throughout the breeding season
(Penfold et al., in press), clear or opalescent semen con-
tained large numbers of epithelial cells and only a few sper-
matozoa of poor motility and high abnormal morphology.
Such ejaculates were typical of April and June, representing
the onset and end of the season, respectively. Eleven (6.1%)
of the ejaculates contained significant amounts of accessory
fluids resulting in low sperm concentrations (< 1 3 109 ml?1)
and eight (4.0%) samples were contaminated with faeces.
These 19 ejaculates were not included in the studies. The
mean semen volume, pH and osmolarity of ejaculates are
shown (Table 1). Only about 15% of collected spermatozoa
were pleiomorphic, with predominant abnormalities includ-
ing a bent mid-piece, head encasement in a cytoplasmic
droplet and microcephaly (Table 1).
Experiment 1: semen characteristics and sperm motility
after ejaculate collection with and without BPSE at
248C or at 48C
There was a significant advantage of diluting freshly
collected spermatozoa 1:1 in BPSE on sperm motility and
progressive status regardless of temperature (248C versus
48C; Table 2). However, further dilution in BPSE (1:10 
or 1:100) caused a transient increase and a rapid decrease
in progressive motility (data not shown). None of the 
treatments influenced the proportions of morphologically
normal spermatozoa (P > 0.05).
Experiment 2: fertility and hatchability of eggs after
artificial insemination with extended semen versus
natural mating
Seasonal commencement of oviposition greatly en-
270 L. M. Penfold et al. 
Table 1. Mean (6 SEM) mid-season ejaculate and sperm traits for
20 representative Northern pintail ducks (Anas acuta)
Ejaculate volume (m l) 66.0 6 14.1
Semen pH 8.5 6 0.1
Semen (undiluted) osmolarity (mOsm) 275 6 8.2
Sperm concentration (3 109 ml?1) 5.7 6 4.3
Sperm motility (%) 57.4 6 6.1
Sperm progressive motilitya 2.7 6 0.2
Structurally normal spermatozoa (%) 85.6 6 1.7
Structurally abnormal spermatozoa (%)
Bent head/midpiece 7.1 6 1.1
Head encased in cytoplasmic droplet 3.3 6 0.7
Microcephalic 3.2 6 0.7
Giant cell 0.5 6 0.2
Bent flagellum 0.2 6 0.2
aOn a scale of 0?5, with 0 indicating no forward progression and 5 indicating
rapid, linear movement.
hanced introduction of the AI catheter into the female
tract. Insemination with 100 m l of extended semen con-
taining 0.31 3 109 motile spermatozoa (number of sper-
matozoa in 100 m l 3 percentage motility) resulted in
fertilization and hatch rates equivalent to those after a nat-
ural mating (Table 3). In contrast, an AI dose of 200 m l
(0.62 3 109) decreased egg fertilization by 40% and
hatchability by > 50% (Table 3). With the exception of the
total number of motile spermatozoa inseminated, all other
sperm parameters were similar (Table 3). When data were
examined across individual birds, it was discovered that a
single insemination (100 m l) of 0.31 6 0.1 3 109 motile
spermatozoa was sufficient to fertilize up to four eggs. 
Experiment 3: fertility and hatchability of eggs after
artificial insemination with 100 m l of extended cold-
stored semen
Cold storage through 72 h had no impact on normal
sperm morphology or viability (Table 4). Sperm motility
and progressive motility were unaffected through 48 h, but
decreased after 72 h (P < 0.05; Table 4). Sperm concentra-
tion did not decrease over time and was similar (P > 0.05)
for the four time storage groups (Table 5). However, hens
in the 72 h group received lower total numbers of motile
spermatozoa (0.11 3 109) than hens in the 0 (0.14 3 109),
24 (0.15 3 109) or 48 h (0.14 3 109) groups owing to the
decrease in sperm motility (Table 4). Nonetheless, live off-
spring were produced from eggs fertilized with extended
semen cold-stored for up to 72 h (Table 5). The incidence
of fertilization using spermatozoa stored for 72 h was only
half that of the 0 and 24 h groups. One hen in the 0 h
group failed to produce fertile eggs compared with two
hens in the 24 h and 48 h groups each and seven hens in
the 72 h group. Among birds within the groups, fertiliza-
tion success ranged from 0 to 100%, 0 to 90%, 0 to 87.5%
and 0 to 75% for the 0, 24, 48 and 72 h groups, respec-
tively. However, owing to individual variation among
birds, there were no differences (P > 0.05) in fertilization
success among the four time storage groups. Neither was
there an effect (P > 0.05) of storage interval on incident of
embryo loss or subsequent hatching success (Table 5).
Experiment 4: impact of freezing and thawing methods
on sperm motility and viability
For all assessments, the number of viable (plasma mem-
brane-intact) spermatozoa was equivalent to the number
of motile spermatozoa. Only motility data are presented
(Table 6). Cryopreserving duck spermatozoa in the ab-
sence of cryoprotectant caused complete (100%) loss of
motility and viability (Table 6). Concentrations of 1%
DMSO and 1% glycerol were also ineffective in maintain-
ing post-thaw motility and viability, with 0% of spermato-
zoa surviving the freeze?thaw process. Adding DMSO or
glycerol (4%) had a variable effect, reducing (P < 0.05)
pre-freeze motility in two of three trials compared with
controls (no cryoprotectant; Table 6). The highest numbers
(P < 0.05) of motile and viable spermatozoa were recov-
ered after the slow?fast freezing and slow thawing with ei-
ther 4% DMSO (32 6 8.3%) or 4% glycerol (35 6 0.0%;
Table 6). Slow thawing consistently resulted in higher
(P < 0.05) numbers of motile spermatozoa after the fast
and slow?fast freezing methods, but not (P > 0.05) after
the slow freezing method. The number of spermatozoa re-
taining motility or viability 24 h after thawing was low
(0?10%), precluding AI trials. 
Discussion
The present study describes normative ejaculate traits and
the sensitivity of duck spermatozoa to cold storage, cry-
oprotectant concentration and freeze?thaw methods.
There were five primary findings: (1) excellent quality
ejaculates containing high proportions of motile and vi-
able spermatozoa could be collected consistently from in-
dividual drakes within ~20 s by simple massage; (2) duck
sperm motility was enhanced after semen collection di-
rectly into BPSE, but was unaffected by short-term transfer
to the laboratory at 24 versus 48C; (3) fertility after AI with
extended fresh semen was similar to that of natural mating;
(4) live offspring could be produced after AI with extended
semen that had been cold-stored for up to 72 h; and (5)
sperm survival after cryopreservation was cryoprotectant-
and freeze?thaw rate-dependent. The use of 4% DMSO or
glycerol combined with cooling at 18C min?1 to ?208C and
then 608C min?1 to ?708C before LN2 immersion resulted
in 35% post-thaw sperm motility. However, post-thaw
motility was short-lived.
Obtaining a clean ejaculate is essential for securing viable
spermatozoa that can withstand storage while retaining
fertilizing capacity. Watery fluid or faecal contamination
Preservation of waterfowl spermatozoa 271
Table 2. Effect of collecting Northern pintail (Anas acuta) ejaculate with or without Beltsville poultry sexton extender (BPSE) at ambient
temperature or 48C (mean 6 SEM)
Treatment Number of ejaculates Sperm motility (%) Sperm progressive motility* Normal sperm morphology (%)
Raw 20 56.5 ? 6.4a 2.7 ? 0.2a 69.5 ? 4.9a
BPSE 248C 20 72.0 ? 4.1b 3.6 ? 0.1b 75.3 ? 3.8a
BPSE 48C 20 79.3 ? 4.0b 3.7 ? 0.1b 72.7 ? 2.9a
*On a scale of 0?5, with 0 indicating no forward progression and 5 indicating rapid, linear movement.
abColumn means with different superscripts differ significantly (P < 0.05).
markedly reduce sperm longevity and fertilizing capacity
in birds (Lake, 1971; Fujihara and Nishiyama, 1976).
Excellent quality, uncontaminated ejaculates were col-
lected routinely by a simple massage method. This has not
been the case in other species such as the Muscovy duck
(Cairina moshata) in which semen collection requires elec-
troejaculation or an artificial vagina?intercept method
(Tan, 1980; Watanabe et al., 1981; Maeda et al., 1984).
Spermatozoa also have been collected by massage and by
electroejaculation from the Pekin duck (Chelmonska et al.,
1962; Pingel, 1972). 
Waterfowl have a well-developed, spirally twisted cop-
ulatory organ, usually coiled within the ventral cloacal
cavity. During mating, this organ uncoils to carry semen
along numerous small crevices, presumably ensuring
semen distribution within the female tract. Eversion of the
copulatory organ during semen collection should be
avoided to prevent the viscous semen spreading over the
irregular surface and into the small crevices, impairing
semen recovery. 
Northern pintail drakes produced a wide range of semen
volumes (10?550 m l), similar to that of the domestic goose
(10?600 m l; Johnson, 1954), but averaging 66 m l, which
was less than the mean (390 m l) in Muscovy duck ejacu-
lates (Tan, 1980). Northern pintail sperm concentration
(5.7 3 109) was higher than that reported for the mallard
(1.3 3 109; Stunden et al., 1998) and less than that of the
Muscovy duck (9.5 3 109; Tan, 1980). Semen from most
avian species has a near neutral pH (Pekin duck, 7.3; mal-
lard duck, 6.8; domestic fowl, 7.0?7.3; Gee, 1983). Of the
avian species studied to date, the Northern pintail appears
to have the most basic semen of pH 8.5, slightly higher
than that of the whooping crane (Grus americana; pH 8.0).
The general morphology of individual Northern pintail
spermatozoa was consistent with observations made in
other Anseriformes (McFarlane, 1962). The Northern 
pintail ejaculate contained low percentages of abnormal
spermatozoa, including those types already described in
other avian species (bent mid-piece, swollen head, giant
sperm and head encased in a cytoplasmic droplet; Gee,
272 L. M. Penfold et al. 
Table 4. Northern pintail (Anas acuta) sperm traits after cold storage (48C) over time
Time (h) Sperm motility (%) Sperm progressive motility* Normal sperm morphology (%) Sperm viability (%)
0 83.1 6 2.3a 3.8 6 0.2a 92.2 6 0.7a 77.6 6 5.2a
24 80.4 6 2.4a 3.7 6 0.2a 90.2 6 1.0a 79.0 6 5.9a
48 70.0 6 3.3a 3.2 6 0.2a 89.9 6 0.8a 76.1 6 4.6a
72 63.9 6 4.0b 2.9 6 0.2b 89.3 6 0.7a 76.7 6 4.4a
*On a scale of 0?5, with 0 indicating no forward progression and 5 indicating rapid, linear movement.
abColumn means with different superscripts differ significantly (P < 0.05).
Table 5. Mean (6 SEM) fertilization and hatch success of Northern pintail (Anas acuta) eggs after artificial insemination with 100 m l of
extended semen cold stored (48C) for 0, 24, 48 or 72 h
Sperm concentration 
Storage time (h) Number of eggs (3 109 ml?1) Fertilized eggs (%) Embryo loss (%) Hatched eggs (%)
0 50 1.7 6 0.3 51.6 6 19.9 9.8 6 7.3 38.0 6 18.3
24 40 1.9 6 0.2 51.5 6 16.5 21.0 6 13.9 30.5 6 12.3
48 56 2.0 6 0.2 41.1 6 10.9 1.3 6 1.3 35.9 6 10.6
72 67 1.7 6 0.2 22.3 6 7.6 2.8 6 1.5 19.5 6 2.8
There were no significant differences.
Table 3. Mean (6 SEM) fertilization and hatch success of Northern pintail (Anas acuta) eggs after natural mating or artificial insemination
with 100 or 200 m l freshly collected, cooled (48C) and extended semen
Insemination Number of Sperm concentration Sperm motility Fertilized eggs Hatched eggs 
method eggs (3 109 ml?1) (%) (%) (%)
Natural mating 14 n/a n/a 100 6 0.0a 100 6 0.0a
AI (100 m l) 14 3.7 6 1.1 84.4 6 4.4 100 6 0.0a 100 6 0.0a
AI (200 m l) 67 3.7 6 1.1 84.4 6 4.4 60.6 6 10.0b 47.6 6 10.0b
abColumn means with different superscripts differ significantly (P < 0.05).
1983), as well as microcephaly. In a parallel study that ex-
amined sperm quality throughout the breeding season
(Penfold et al., in press) the head encased in a cytoplasmic
droplet malformation was found to be most prevalent near
the end of the season. The bent midpiece defect was char-
acterized by acute flexion in the midpiece region so that
the head was parallel to the flagellum. This common de-
fect occurs in diluted and stored chicken and turkey sper-
matozoa, probably as a result of osmotic stress (Bakst and
Sexton, 1979; Bakst, 1990). The fowl spermatozoa plasma
membrane in this region is known to be acutely suscepti-
ble to stress (Yamane et al., 1966). In the present study, it
was likely that the bent midpiece in > 7% of the spermato-
zoa was not osmotically induced by adding BPSE because
the osmolarity was adjusted to that of the neat ejaculate.
Furthermore, BPSE extension did not increase the propor-
tion of abnormal spermatozoa compared with the neat
ejaculate. 
Short-term (24 h) cold-storage of spermatozoa has been
reported for domestic fowl spermatozoa, with a fertility
success of 72% (versus 90% for fresh spermatozoa;
Blesbois and de Reviers, 1992). Similarly, AI using short-
term (24 h) cold-stored turkey semen resulted in a fertility
rate of 61% (versus 91% for unstored controls; Sexton,
1980a). Although Northern pintail sperm motility de-
creased after 72 h, no change was apparent in other sperm
traits. This finding was in contrast to another study of do-
mestic drake semen (Kasai et al., 1996) that showed a 33%
decrease in viability after 24 h cold storage, assessed using
the same staining combination used in this study. In the
study of Kasai et al. (1986), semen was diluted 1:5 in
Lake?s diluent before storage (compared with 1:3 in BPSE
in the present study) so a dilution effect (Mann, 1964) 
may have been partially responsible for this different re-
sult. In the present study, sperm concentration did not de-
crease through 72 h of storage. This is significant because
avian sperm death often is accompanied by cell lysis 
and complete cell disintegration; for example, the rate 
of turkey sperm lysis can range from 17 to 33% of the 
unstored controls (Sexton, 1988b). Accordingly, as dead
spermatozoa disintegrate in the ejaculate, the percentage
sperm motility and viability may appear to remain 
unchanged over time.
AI with extended fresh semen achieved fertility and
hatch success similar to those of natural breeding under
captive conditions (100%). Hatch success in wild
Northern pintails is reported to be 32?68% (del Hoyo et
al., 1992) owing to nest predation or abandonment. Thus,
these results illustrate the potential of AI for improving
overall reproductive efficiency in waterfowl species.
Although copulation frequency during natural breeding re-
mains unknown for the Northern pintail, the results 
revealed that insemination of 0.31 3 109 (motile) sperma-
tozoa (Expt 2) was sufficient to fertilize at least four con-
secutively laid eggs. The lower fertilization success in Expt
3 compared with Expt 2 was probably related to insemi-
nating numbers of spermatozoa below that required to
achieve maximal fertilization. It is known that the sperm
storage crypts of hens need to be adequately filled for opti-
mal fertilization success (Brillard, 1993). As the 200 m l
(0.63 3 109 spermatozoa) inseminations in Expt 2 failed to
result in maximal fertilization, the optimal AI dose for the
Northern pintail appears to be about 0.31 3 109 motile
spermatozoa but over-insemination can be detrimental to
fertility. The smaller number of spermatozoa inseminated
in Expt 3 compared with Expt 2 at 0 h may account for the
difference in fertilization success between the two experi-
ments. In addition, the high degree of variability among fe-
males in Expt 3 made it difficult to demonstrate the loss of
fertility in the cold-stored semen. Although there was no
statistical effect through 72 h of storage on fertilization and
hatch success, the range of fertilized eggs in the treatments
groups (0?100%, 0?90%, 0?87.5% and 0?75% for the 0,
24, 48 and 72 h groups, respectively) indicated a trend in
Preservation of waterfowl spermatozoa 273
Table 6. Influence of sperm freezing and thawing method and cryoprotectant on Northern pintail
(Anas acuta) sperm motility
Freezing method Thawing method Control (0%) Dimethylsulphoxide (4%) Glycerol (4%)
Fast Prefreeze 55 6 2.9a 58 6 8.3a 53 6 4.4a
Fast 0 6 0.0b 4 6 1.3b 5 6 0.0b
Slow 0 6 0.0b 20 6 10c 22 6 6.7c
Slow?fast Prefreeze 73 6 4.4b 58 6 1.7a 62 6 3.3a
Fast 0 6 0.0d 15 6 12c 7 6 6.3c
Slow 0 6 0.0d 32 6 8.3e 35 6 0.0e
Slow Prefreeze 63 6 12b 43 6 7.5a 58 6 7.5ab
Fast 0 6 0.0c 1 6 0.0c 1 6 0.5c
Slow 0 6 0.0c 3 6 2.0c 2 6 1.0c
abcdeColumn and row means within freezing method with different superscripts differ significantly (P < 0.05).
fertility reduction that was probably masked by suboptimal
insemination doses. 
Although Northern pintail spermatozoa are remarkably
tolerant to cold storage at 4?C, they were susceptible to
the stresses associated with cryopreservation and thawing.
Comparative assessments revealed that a combination of
initial slow cooling followed by fast cooling (from 4 to
?208C at a rate of 18C min?1, approximately 608C min?1 to
?708C, immersed in LN2; Gee and Sexton, 1980) was most
effective for recovering motile, viable spermatozoa.
DMSO and glycerol were equally effective in protecting
spermatozoa during freezing and thawing, but only at the
4% concentration. A 1% concentration of either cryopro-
tectant clearly was ineffective. A similar comparative ap-
proach for the sandhill crane and domestic fowl has 
revealed that 4% DMSO is the most effective cryoprotec-
tant for preserving post?thaw sperm motility (Sexton and
Gee, 1978). A standard cryoprotectant for poultry, DMSO
also has the advantage of being safe when inseminated
into females (Sexton, 1980b; Lake and Ravie, 1984). In
contrast, glycerol has a contraceptive effect when used in
a similar way (Lake and Stewart, 1978; Lake et al., 1980)
and so must be diluted from the spermatozoa before AI
(Lake et al., 1980). Northern pintail spermatozoa tolerated
cryopreservation rates of approximately 608C min?1, even
though avian sperm generally must be cooled slowly (at
approximately 18C min?1; Sexton and Gee, 1978; Gee,
1983). Freezing in LN2 vapour has been reported for
Muscovy spermatozoa by Watanabe et al. (1981), but no
freezing rate was described. The LN2 technique requires
further exploration because its simplicity could permit use
under field conditions. Because sperm motility was not
sustained beyond 24 h of thawing, it was not possible to
determine the fertility of thawed spermatozoa in this 
experiment and more research is required to develop ef-
fective cryopreservation of Northern pintail duck sperma-
tozoa. Lastly, and unsurprisingly, preliminary results (data
not shown) indicated that post-thaw sperm viability is
linked with pre-freeze sperm quality, and that good quality
ejaculates resulted in more spermatozoa surviving the
freeze?thaw process.
These studies provide a foundation for the collection
and short-term storage of semen from waterfowl species.
Although the cryopreservation of non-domestic waterfowl
spermatozoa requires more work, cold storage (48C) of 
waterfowl semen appears to be a suitable alternative in the
interim, providing adequate fertility for up to 3 days after
collection. This time frame is sufficient to allow national
and international shipment of semen from valuable water-
fowl for conservation or commercial purposes.
This work was supported by the Philip Reed Foundation and
Friends of the National Zoo. The authors are grateful to George
Gee, Steve Leathery and Denise Holsberger for helpful advice 
on semen collection and AI, the CRC Bird Unit for care and 
assistance and Francisco Palmer and Tanya Herzog for excellent
technical assistance.
References
Ankney CD, Dennis DG, Wishard LN and Seeb JE (1986) Low genetic
variation between black ducks and mallards Auk 103 701?709
Bakst MR (1990) Preservation of avian cells. In Poultry Breeding and
Genetics pp 91?108 Ed. RD Crawford. Elsevier Science Publishers,
Amsterdam
Bakst MR and Sexton TS (1979) Fertilizing capacity and ultrastructure of
fowl and turkey spermatozoa before and after freezing Journal of
Reproduction and Fertility 60 121?127
Blesbois E and de Reviers M (1992) Effect of different fractions of seminal
plasma on the fertilizing ability of fowl spermatozoa stored in vitro.
Journal of Reproduction and Fertility 95 263?268 
Brillard JP (1993) Sperm storage and transport following natural mating
and artificial insemination Poultry Science 72 923?928
Burrows WH and Quinn JP (1939) Artificial insemination of chickens and
turkeys US Department of Agriculture Circular 525 12
Chelmonska B, Galuska H and Lisiecki J (1962) Electroejaculation in
drakes Medycyna weterynaryjna 18 712?714
del Hoyo J, Elliott A and Sargatal J (1992) Anseriformes. In Handbook of
the Birds of the World Vol. 1 p. 555 Eds J del Hoyo, A Elliott and J
Sargatal. Lynx Edicions, Barcelona
Ellis-Joseph S, Hewston N and Green A (1992) In Global Waterfowl
Conservation Assessment Management Plan p 2. Captive Breeding
Specialist Group, Apple Valley, MN
Fujihara N and Nishiyama H (1976) Studies on the accessory reproductive
organs in the drake 5. Effects of the fluid from the ejaculatory 
groove region on the spermatozoa of the drake Poultry Science 55
2415?2420
Garner DL, Johnson LA, Yue ST, Roth BL and Haughland RP (1994) Dual
DNA staining assessment of bovine sperm viability using SYBR-14 and
propidium iodide Journal of Andrology 15 620?629
Gee G (1983) Avian artificial insemination and semen preservation. In
IFCB Symposium on Breeding Birds in Captivity pp 375?398 Eds AC
Risser, Jr and FS Todd. International Foundation for the Conservation
of Birds, North Hollywood, CA
Gee G and Sexton TJ (1980) Artificial insemination and semen preserva-
tion of nondomestic birds Proceedings American Association of Zoo
Veterinarians pp 55?58
Gee G and Sexton TJ (1990) Cryogenic preservation of semen from the
Aleutian Canada Goose (Branta canadensis leucopareia) Zoo Biology 9
361?371
Hediger H (1965) Environmental factors influencing the reproduction of
zoo animals. In Sex and Behavior pp 319?354 Ed. FA Beach. John
Wiley and Sons, New York
Howard J, Bush M and Wildt DE (1991) Teratospermia in domestic cats
compromises penetration of zona-free hamster ova and cat zonae 
pellucidae Journal of Andrology 12 36?45
Johnson AS (1954) Artificial insemination and duration of fertility in geese
Poultry Science 39 230?232
Kasai K, Izumo A and Nishimura K (1996) Evaluation of sperm viability in
fresh and stored drake semen using dual fluorescent staining
Proceedings of Satellite Symposium in the 8th American Association of
Avian Pathologists Kinki University, Wakayama, Japan, 21?22
Lake PE (1971) The male in reproduction. In Physiology and Biochemistry
of the Domestic Fowl pp 1411?1447 Eds DJ Ball and BM Freeman.
Academic Press, New York
Lake PE and Ravie O (1984) An exploration of cryoprotective compounds
for fowl spermatozoa British Poultry Science 25 145?150
Lake PE and Stewart JM (1978) Preservation of fowl semen in liquid nitro-
gen, an improved method British Poultry Science 19 187?194
Lake PE, Buckland RB and Ravie O (1980) Effect of glycerol on the viabil-
ity of fowl spermatozoa, implications for its use when freezing semen.
Cryoletters 1 299?304
McFarlane RW (1962) The Taxonomic Significance of Avian Sperm MSc
Thesis, University of Florida
Maeda T, Terada T and Tsutsumi Y (1984) Morphological observations on
frozen and thawed Muscovy spermatozoa British Poultry Science 25
409?413 
274 L. M. Penfold et al. 
Mann T (1964) The Biochemistry of Semen and the Male Reproductive
Tract John Wiley and Sons, New York
Nei M, Maruyama T and Chakraborty R (1975) The bottleneck effect and
genetic variability in populations Evolution 29 1?10
Penfold LM, Wildt DE, Herzog TL, Lynch W, Ware, L, Derrickson SE and
Monfort SL Seasonal patterns of LH and testosterone and semen qual-
ity in the Northern pintail (Anas acuta) (in press)
Pingel M (1972) A contribution to artificial insemination in duck breeding.
Wissenschaftliche Zeitschrift der Humboldt Universitat zu Berlin
Mathematische-Naturwissenschatliche Reihe 21 179?181
Polge C (1951) Functional survival of fowl spermatozoa after freezing at
?798C Nature 331 70?72
Ralls K and Ballou JD (1986) Preface to the proceedings of the work-
shop on genetic management of captive populations Zoo Biology 5
81?86
Sexton TJ (1977) A new poultry semen extender 1. Effect of extension on
the fertility of chicken semen Poultry Science 56 1443?1446
Sexton TJ (1980a) Recent advances in semen storage of the fowl and
turkey Proceedings of The 9th International Congress of Animal
Reproduction and Artificial Insemination (Madrid) 527?533
Sexton TJ (1980b) A new poultry semen extender 5. Relationship of dilu-
ent components to cytotoxic effects of dimethylsulphoxide on turkey
spermatozoa Poultry Science 59 1142?1144
Sexton TJ (1981) Development of a commercial method for freezing 
turkey semen I. Effect of pre-freeze techniques on the fertility of pro-
cessed unfrozen and frozen?thawed semen Poultry Science 60
1567?1573
Sexton TJ (1988a) Research note: influence of damaged spermatozoa on
the fertility of turkey semen stored for 24 hours at 58C Poultry Science
67 1483?1485
Sexton TJ (1988b) Comparison of commercial diluents for holding turkey
semen 24 hours at 58C Poultry Science 67 131?134
Sexton TJ and Gee G (1978) A comparative study on the cryogenic 
preservation of semen from the sandhill crane and the domestic fowl
Symposium of the Zoological Society of London 43 89?95
Stunden CE, Bluhm CK, Cheng KM and Rajamahendran R (1998) Plasma
testosterone profiles, semen characteristics and artificial insemination
in yearling and adult captive mallard ducks (Anas platyrhynchos)
Poultry Science 77 882?887 
Systat Inc. (1989) Systat for the Macintosh Version 5.2 Evanston, IL
Tai JJ, Su YM, Huang HH, Terada T and Watanabe M (1983) Studies on
the cryopreservation of drake semen. New strategies for improving 
animal production for human welfare Proceedings of the 5th World
Conference on Animal Production pp 183?184
Tan NS (1980) The frequency of collection and semen production in
Muscovy ducks British Poultry Science 21 265?272
Thurston RJ (1995) Storage of poultry semen above freezing for twenty-
four to forty-eight hours Proceedings of the 1st International
Symposium on the Artificial Insemination of Poultry?International
Symposium on the Artificial Insemination of Poultry pp 107?122 
Eds MR Bakst and GJ Wishard. Poultry Science Association, Savoy, IL
Tselutin K, Narubina L, Mavrodina T and Tur B (1995) Cryopreservation
of poultry semen British Poultry Science 36 805?811
US Fish and Wildlife Service (1999) Status of Ducks Office of Migratory
Birds Annual Report
Watanabe M, Matsumoto Y, Takeshita N and Terada T (1981) Fertility of
Muscovy semen frozen for about three years Journal of the Faculty of
Applied Biological Science 20 81?85
Yamane JS, Tsukunaga S and Takahashi T (1966) ?Hiroshima? method of
artificial insemination of the domestic fowl Journal of Faculty of Fish
and Animal Husbandry 6 395?429 
Received 19 April 2000.
Accepted 23 August 2000.
Preservation of waterfowl spermatozoa 275