Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
ABSTRACT
This study investigated the hypothesis that dairy
heifers divergent in genetic merit for fertility traits dif-
fer in the age of puberty and reproductive performance.
New Zealand’s fertility breeding value (FertBV) is the
proportion of a sire’s daughters expected to calve in
the first 42 d of the seasonal calving period. We used
the New Zealand national dairy database to identify
and select Holstein-Friesian dams with either positive
(POS, +5 FertBV, n = 1,334) or negative FertBV
(NEG, −5% FertBV, n = 1,662) for insemination with
semen from POS or NEG FertBV sires, respectively.
The resulting POS and NEG heifers were predicted to
have a difference in average FertBV of 10 percentage
points. We enrolled 640 heifer calves (POS, n = 324;
NEG, n = 316) at 9 d ± 5.4 d (± standard deviation;
SD) for the POS calves and 8 d ± 4.4 d old for the
NEG calves. Of these, 275 POS and 248 NEG heif-
ers were DNA parent verified and retained for further
study. The average FertBV was +5.0% (SD = 0.74)
and −5.1% (SD = 1.36) for POS and NEG groups,
respectively. Heifers were reared at 2 successive facili-
ties as follows: (1) calf rearing (enrollment to ~13 wk of
age) and (2) grazier, after 13 wk until 22 mo of age. All
heifers wore a collar with an activity sensor to monitor
estrus events starting at 8 mo of age, and we collected
weekly blood samples when individual heifers reached
190 kg of body weight (BW) to measure plasma pro-
gesterone concentrations. Puberty was characterized
by plasma progesterone concentrations >1 ng/mL in
at least 2 of 3 successive weeks. Date of puberty was
defined when the first of these samples was >1 ng/mL.
Heifers were seasonally bred for 98 d starting at ~14
mo of age. Transrectal ultrasound was used to confirm
pregnancy and combined with activity data to estimate
breeding and pregnancy dates. We measured BW every
2 wk, and body condition and stature at 6, 9, 12, and
15 mo of age. The significant FertBV by day interaction
for BW was such that the NEG heifers had increas-
ingly greater BW with age. This difference was mir-
rored with the significant FertBV by month interaction
for average daily gain, with the NEG heifers having a
greater average daily gain between 9 and 18 mo of age.
There was no difference in heifer stature between the
POS and NEG heifers. The POS heifers were younger
and lighter at puberty, and were at a lesser mature
BW, compared with the NEG heifers. As a result, 94 ±
1.6% of the POS and 82 ± 3.2% of the NEG heifers had
reached puberty at the start of breeding. The POS heif-
ers were 20% and 11% more likely to be pregnant after
21 d and 42 d of breeding than NEG heifers (relative
risk = 1.20, 95% confidence interval of 1.03–1.34; rela-
tive risk = 1.11, 95% confidence interval of 1.01–1.16).
Results from this experiment support an association
between extremes in genetic merit for fertility base on
cow traits and heifer reproduction. Our results indicate
that heifer puberty and pregnancy rates are affected
by genetic merit for fertility traits, and these may be
useful phenotypes for genetic selection.
Key words: puberty, pregnancy, heifer, genetic,
fertility
Heifers with positive genetic merit for fertility traits reach puberty
earlier and have a greater pregnancy rate than heifers
with negative genetic merit for fertility traits
S. Meier,1* L. R. McNaughton,2 R. Handcock,2† P. R. Amer,3 P. R. Beatson,4 J. R. Bryant,1,5‡
K. G. Dodds,6 R. Spelman,2 J. R. Roche,1§ and C. R. Burke1
1DairyNZ Limited, Private Bag 3221, Hamilton 3240, New Zealand
2Livestock Improvement Corporation, Hamilton 3240, New Zealand
3AbacusBio Limited, Dunedin 9016, New Zealand
4CRV-Ambreed, Hamilton 3216, New Zealand
5New Zealand Animal Evaluation Limited, Private Bag 3221, Hamilton 3240, New Zealand
6AgResearch, Invermay, Agricultural Centre, Private Bag 50034, Mosgiel 9053, New Zealand
J. Dairy Sci. 104
https://doi.org/10.3168/jds.2020-19155
© 2021, The Authors. Published by Elsevier Inc. and Fass Inc. on behalf of the American Dairy Science Association®.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Received June 24, 2020.
Accepted October 19, 2020.
*Corresponding author: Susanne.Meier@ dairynz .co .nz
†Current affiliation: Institute of Veterinary, Animal and Biomedical
Sciences, Massey University, Private Bag 11-222, Palmerston North
4474, New Zealand.
‡Current affiliation: AgResearch, Ruakura Agricultural Centre, 10
Bisley Rd., Hamilton 3214, New Zealand.
§Current affiliation: Ministry for Primary Industries-Manatū Ahu
Matua, Charles Ferguson Tower, Pipitea, Wellington 6140, New
Zealand, and University of Auckland, School of Biological Sciences,
University of Auckland, Private Bag 92019, Auckland 1142, New
Zealand.
Journal of Dairy Science Vol. 104 No. 3, 2021
INTRODUCTION
Before 2000, both the phenotypic reproductive per-
formance and the genetic merit for fertility traits of
lactating dairy cows were declining (Berry et al., 2014;
Pryce et al., 2014). This decline led many dairy genetics
organizations to extend breeding objectives to include
fertility traits (Miglior, 2002; Miglior et al., 2005; Har-
ris et al., 2006; Egger-Danner et al., 2015). As a result,
genetic merit for fertility traits has been improving,
with albeit modest yearly gains estimated at <0.2 per-
centage points (Pryce et al., 2014).
Under traditional, pedigree-based genetic evaluation
approaches, the rate of genetic gain in fertility can be
accelerated by increasing the accuracy and volume of
phenotypic data, as well as finding novel and earlier
genetically correlated traits from the progeny of sires.
Opportunities such as increasing the accuracy of ex-
isting industry data may be implemented quickly, but
promise only modest gains. Whereas, incorporation
of new fertility traits that can be evaluated earlier or
that have a greater heritability are expected to produce
greater gains (Berry et al., 2014; Carthy et al., 2014;
Bowley et al., 2015; Jenkins et al., 2016). For these
reasons, interest in the evaluation of novel fertility
traits such as resumption of cycling postpartum, estrus
behaviors, and pregnancy loss is increasing (Petersson
et al., 2007; Bamber et al., 2009; Berry et al., 2014;
Fleming et al., 2015; Lucy, 2019). These novel measures
may be the next generation of traits that increase ge-
netic gain in fertility, leading to further improvements
in cow reproductive performance.
A strong candidate trait that increases the rate of
genetic gain in fertility should be determined earlier
than those in current use, have a heritability greater
than current traits, and be positively correlated with
the key outcomes being selected for, such as pregnancy
rate. Traits of interest could include the age at pu-
berty and heifer pregnancy rate. These candidate traits
are measured before calving-related trait and have a
greater heritability than traditional reproductive traits
captured after calving (Morris et al., 2000, 2011). Ad-
ditionally, phenotypic and genetic correlations between
heifer traits and subsequent fertility suggest that heif-
ers that calve early have better fertility as cows (Pryce
et al., 2007; Tiezzi et al., 2012), indicating that selec-
tion for heifer fertility traits could result in better cow
fertility. Additionally, Wathes et al. (2014) identified
a positive relationship between heifer reproductive
performance (e.g., age at first calving) and subsequent
calving interval. Yet, other studies have reported weak
or no genetic or phenotypic association between heif-
ers and cow fertility (Raheja et al., 1989; Mion et al.,
2019).
To identify candidate traits that accelerate the rate
of genetic gain for fertility, we wanted to understand
the underlying biological differences between heifers
with high and low values for New Zealand’s fertil-
ity breeding value (FertBV). We generated a unique
population of genetically divergent animals that rep-
resented a research resource to support the evaluation
of traditional and novel measures related to cow fertil-
ity and reproduction. Previous research investigated
phenotypic differences of cows that were divergent in
genetic merit for fertility traits, identifying differences
in the timing of conception, conception, luteal and fol-
licular function, uterine health, and the somatotropic
axis in cows with positive or negative genetic merit
for fertility traits (Cummins et al., 2012a,b,c; Moore
et al., 2014). Another study identified heifers based on
genomic selection for heifer conception rate and daugh-
ter pregnancy rates in the United States, with these
authors reporting a range of heifer traits (Veronese et
al., 2019a,b). In the current study, we report evalua-
tions on heifer calves through to their first successful
breeding period. Specifically, we hypothesized that the
onset of puberty and reproductive performance would
differ for heifers of high and low FertBV. A secondary
hypothesis was that no difference in the growth and
development of the heifers with high and low FertBV
would be observed. To test these hypotheses, we mea-
sured the age at puberty, the pregnancy outcomes dur-
ing their first breeding season, and the heifers’ growth
and development.
MATERIALS AND METHODS
The Ruakura Animal Ethics Committee (Hamilton,
New Zealand) approved this study and all manipula-
tions (AE application #13574).
Establishing the Research Herd
The process for dam selection, contract breeding, and
calf collection is depicted in Supplemental Figure S1
(https: / / data .mendeley .com/ datasets/ 343t97cpdr/ 1)
and described herein.
Breeding Strategy. We used a customized, seasonal
breeding strategy between October and November 2014
to produce heifer calves with a predicted high (POS,
+5%) and low (NEG, −5%) genetic merit for fertility
traits via assortative breeding between parents with
POS and NEG estimated FertBV as defined by the New
Zealand national genetic evaluation scheme (evaluation
run Feb 2014). The FertBV is expressed as the percent-
age of a sire’s daughters that are predicted to calve in
the first 42 d of the calving season. Therefore, a FertBV
of +5 equates to 5% more daughters calving in the first
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
42 d of the calving season, and −5 equates to 5% fewer
daughters calving in the first 42 d compared with a 0
FertBV. At the time, the FertBV was estimated from
8 predictor traits as follows: presented for breeding
within 21 d of the planned start of seasonal breeding in
lactations 1, 2 and 3; recalving in the first 42 d after the
planned start of seasonal calving in lactations 2, 3, and
4; and milk volume and BCS at 60 DIM in the cow’s
first lactation (DairyNZ, 2018).
We generated a breeding plan that was targeted to
produce the desired difference in the FertBV of the
offspring, and for the criteria set using the MateSel
(Kinghorn, 2011). This approach modeled the mating
outcomes and produced a customized breeding strategy
that could achieve the 10-percentage point difference in
FertBV in the offspring. Additionally, the customized
breeding strategy was designed to limit the inbreeding
coefficient of the offspring with the average inbreeding
coefficient of 2.8 ± 1.44% (mean ± SD, target <6.3%).
It also aimed to limit the expected parent averages
for milk volume breeding value (BV), fat BV, pro-
tein BV, BW BV (DairyNZ, 2018), and ancestry (%
North American Holstein-Friesian; HF) to be within
1 standard deviation (SD) of each other, and to pro-
duce calves of >15/16th HF. The predicted mean BV,
the SD, and commentary on achievement of predicted
traits for the heifer offspring are summarized in Supple-
mental Table S1 (https: / / data .mendeley .com/ datasets/
343t97cpdr/ 1).
Dam Selection. Suitable dams were selected from
the New Zealand Dairy Industry Good Animal Da-
tabase (https: / / www .dairynz .co .nz/ animal/ animal
-evaluation/ animal -database/ data -access/ ). These
cows had >3 herd tests in the 2012 to 2013 lactational
season, and more than 89% complete pedigree infor-
mation that confirmed the sire and maternal grandsire
were POS or NEG FertBV. The cows were ≥14/16th
HF, had calved in the first 42 d of the seasonal calving
period, with the expectation that this would optimize
reproductive outcomes to the contracted breeding, and
were less than 8 yr old and had a high likelihood of
remaining in the herd. In addition, candidate dams had
no recorded markers for genetic-based diseases. Candi-
date dams were eligible if they came from herds that
were free of tuberculosis, Johne’s disease, and enzootic
bovine leukosis. Herd owners with suitable dams were
enrolled for contracted inseminations. The contracted
inseminations consisted of 1,334 POS and 1,533 NEG
dams from 669 commercial herds and were inseminated
with their respective allocated semen (Supplemen-
tal Figure S1, https: / / data .mendeley .com/ datasets/
343t97cpdr/ 1). Due to the relatively low number of
confirmed breedings and expected calvings of NEG
compared with POS dams (55% vs. 69%; Supplemental
Figure S1), we used the New Zealand Dairy Industry
Good Animal Database to identify additional mating
between NEG sires and NEG dams. Only dams that
had fulfilled the criteria as outlined above were con-
sidered. We identified 129 pregnant NEG dams that
were recorded breeding with a NEG FertBV sire. The
NEG FertBV progeny had an expected birth date in
the same calving season as the other calves. The dams
were identified before calving and enrolled so that they
underwent the same processes precalving. We collected
24 heifer calves from these 129 dams identified. These
were undistinguishable throughout calf collection, calve
rearing, and heifer rearing.
Sire Selection. Sires with POS and NEG FertBV
were selected based on semen availability. Semen from
sires with sufficient stock for 3 inseminations of each
dam was distributed for repeated rounds of insemina-
tions, if required. In total, 24 POS (FertBV 5.1% ±
1.67; mean ± SD) and 43 NEG sires (FertBV −6.1% ±
2.33) were used.
Calf Collection and Parentage Verification.
We obtained 640 female calves from 379 herds during
the 2015 seasonal calving period (Supplemental Figure
S1, https: / / data .mendeley .com/ datasets/ 343t97cpdr/
1). The mean date (± SD) of birth was August 3 ± 14
d (n = 324) for the POS group, and August 7 ± 15 d
(n = 316) for the NEG group. The average age at col-
lection (±SD) was 9 ± 5.4 d for the POS calves and 8
± 4.4 d for the NEG calves. We verified the parentage
of the calf and paternity of their dam via DNA testing
from an ear-notch tissue sample using the commercial
parentage panel available through Genemark (LIC,
Hamilton, New Zealand). Retained calves (POS, n =
289; NEG, n = 276) had a known sire and maternal
grandsire of corresponding genetic merit for fertility
(Supplemental Figure S1, https: / / data .mendeley .com/
datasets/ 343t97cpdr/ 1). The numbers of calves col-
lected per sire are summarized in Supplemental Figure
S2.
Calf and Heifer Rearing
All calves were reared for 13 wk at a single facility
(Parklands Road, Te Awamutu, New Zealand; latitude
−38.018759, longitude 175.440412). On arrival, calves
were placed in indoor pens in groups of 9 calves and
fed 3.5 L of milk replacer once daily (Ancalf, 26%
protein, NZAgbiz, 2020) with commercial calf muesli
(20% protein, SealesWinslow Ltd., Morrinsville, New
Zealand) ad libitum for 7 wk. From wk 8 to 13, calves
were grazed outdoors in groups of 30 to 40 where they
were grazed a predominantly ryegrass (Lolium perenne)
pasture and grass silage, and had access to ad libitum
calf muesli (20% protein, SealesWinslow Ltd.). Heifers
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
were moved to a grazing property at an average age of
95 d (SD = 2.9 d; State Highway 16, Waimauku, New
Zealand; latitude −36.757141, longitude 174.458980).
At the grazing property, the heifers were grouped into
4 age-based herds of 130 to 150 heifers on arrival from
the rearer, with the POS and NEG heifers represented
across grazing herds. Heifers grazed on ryegrass pasture,
with the sward including kikuyu (Pennisetum clandesti-
num) and chicory (Cichorium intybus). Supplementary
feeds (palm kernel expeller and pasture baleage and
silage) were fed to the heifers when insufficient pasture
was available to ensure heifer growth rates were con-
sistent with industry BW targets (DairyNZ, 2016b).
By February 2017, the research herd consisted of 524
heifers (275 POS and 249 NEG). A summary of the
breeding worth, BV for key traits (animal evaluation
run Jan 2017), ancestry, and their respective dams and
sires are presented in Table 1.
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Table 1. Mean (and SD) of breeding worth and component traits of heifers with positive or negative genetic
merit for fertility traits that were available for the reproductive phenotypes including numbers (n), the date
of birth, fertility breeding value (BV), breeding worth, and the components traits, as well as ancestry of the
heifers, and the fertility BV and breeding worth of their dams and sires
Variable per estimated genetic
merit1
Genetic merit for fertility trait
Positive
Negative
Mean SD Mean SD
Heifer (n) 275 — 249 —
Date of birth (d/mo; d) 3 Aug 14 7 Aug 15
Estimated genetic merit
Fertility BV2 (%) 5.0 0.74 −5.1 1.36
Breeding worth3 (NZ$/yr) 109 21.4 40 30.7
Volume BV4 (kg) 654 165.1 732 157.6
Fat BV4 (kg) 11.3 5.46 17.8 6.57
Protein BV4 (kg) 17.8 6.57 23.2 4.57
BW BV4 (kg) 37 12.5 40 10.1
BCS BV50.07 0.068 −0.08 0.071
Gestation length BV (d) −3.2 2.07 −1.4 2.23
Residual survival BV 54 58.1 30 72.7
Total longevity BV (d of life) 300 47.6 74 82.4
SCS BV6−0.11 0.140 0.10 0.175
Ancestry7 (North American %) 56 6.3 62 8.4
Inbreeding coefficient7 (%) 2.6 1.23 3.1 1.62
Dam8 (n) 273 — 246 —
Fertility BV2 (%) 4.6 1.00 −3.6 1.62
Breeding worth3 (NZ$/y) 89 28.7 39 29.9
Sire9 (n) 24 — 43 —
Fertility BV2 (%) 5.1 1.67 −6.1 2.33
Breeding worth3 (NZ$/yr) 132 34.8 35 46.3
1New Zealand Animal Evaluation (NZAEL) animal evaluation run date Jan. 2017.
2Fertility BV is a percentage value consisting of the lactating cow’s ability to start cycling (a binary trait called
PM21, representing success vs. failure at being presented for breeding in the first 21 d of the herd’s breeding
period, from first, second, and third parity cows) and a lactating cow’s ability to conceive (a binary trait called
CR42, representing success vs. failure for recalving in the first 42 d of the herd’s calving period, from second-,
third-, and fourth-parity cows; DairyNZ, 2017).
3Breeding worth (NZ$/yr) is the NZ$ net farm income/5 t of DM, which is assumed to be fed per cow per year.
(At time of writing, US$ equivalent POS BW is US$77 and NEG BW is US$28.)
4The breeding plan aimed to reduce the variation in the BV for milk volume, fat, protein, BW, and ancestry
(% North American Holstein-Friesian) to be within 1 SD and produce calves of >15/16th Holstein-Friesian
breeding.
5Body condition unit is a measure of subcutaneous fat deposits (Roche et al., 2004), calculated using records
collected on primiparous 2-yr-old heifers. These records are collected in early lactation. Raw scores are convert-
ed into a d 60 lactation equivalent, and then enter the animal evaluation model. A breed neutral adjustment
has been applied to this BV, such that the breed average for this trait is 0 across all breeds.
6SCS is the log-transformed SCC, which is derived from milk testing (DairyNZ, 2017).
7Ancestry and inbreeding coefficient data were received from animal evaluation following parentage checks
(Feb. 2016).
8Twin heifer calves were collected from 5 dams.
9Parentage verified sires of the calves. More negative fertility BV were used due to the reduced availability of
semen and to achieve the inbreeding criteria set for the expected offspring.
Journal of Dairy Science Vol. 104 No. 3, 2021
Body Weight, ADG, BCS, and Stature
Average age at first BW measurement was 9 d (SD
= 5.0 d; weighed using static scales, Gallagher, Hamil-
ton, New Zealand). Thereafter, BW was measured once
every 2 wk. Average daily gain was calculated from
the following periods: 9 d to 3 mo, 3 to 6 mo, 6 to 9
mo, 9 to 12 mo, 12 to 15 mo, 15 to18 mo, and 18 to 21
mo of age. Heifer stature and BCS (1–10 scale; Roche
et al., 2004, 2007) were measured at 6, 9, 12, and 15
mo of age. Stature measures were height (vertical dis-
tance from the ground to the top of the withers), girth
(circumference of the animal measured directly behind
the front legs), and length (horizontal distance between
the bottom of the pin bones to the top of the withers;
Macdonald et al., 2007).
Plasma Sampling, Progesterone Analyses,
and Puberty Variables
Weekly blood sampling for determination of plasma
progesterone concentrations started when heifers were
approximately 190 kg of BW (Macdonald et al., 2007)
and continued either until puberty or until 3 wk after
the start of the breeding season for those that had not
reached puberty by this time. Blood was collected from
the coccygeal vessel into evacuated blood tubes contain-
ing lithium heparin (BD Vacutainers, BD New Zealand,
Auckland, New Zealand). Samples were placed in iced
water and transported to the laboratory at the end of
the sampling day and centrifuged (at 4°C, 1,900 × g for
12 min) for plasma harvest. Plasma was stored in dupli-
cate aliquots at −20°C until analysis for progesterone.
A commercial double antibody radioimmunoassay kit
was used to determine plasma progesterone concentra-
tions in accordance with the manufacturer’s instruc-
tions (ImmuChem Progesterone Double Antibody RIA,
MP Biomedicals LLC, Irvine, CA). The inter- and
intra-assay coefficients of variation for a high standard
were 8% and 8%, respectively, and for the low standard
they were 14% and 10%, respectively (n = 25 assays).
The minimal detectable concentration was 0.18 ng/mL.
Puberty was defined to have occurred when proges-
terone concentrations were >1 ng/mL in at least 2 of 3
consecutive weekly plasma samples (Macdonald et al.,
2007). Date of puberty was the day when the first of
these samples was >1 ng/mL. Age at puberty (d) and
estimated BW at puberty were calculated. Estimated
BW at puberty was calculated using the BW, the ADG,
and the age at puberty. Percentage of expected mature
BW at puberty was calculated using the estimated ma-
ture cow BW using the industry standard estimate for
HF cows plus the genetic merit for BW (BW BV) for
that individual (DairyNZ, 2016a).
Heifers (n = 15; 14 NEG, 1 POS) that had not
reached puberty based on the plasma progesterone
concentrations by 21 d after the start of seasonal
breeding underwent transrectal ultrasonography with
a 5 to 15 MHz probe (SonoScape S6V, Euromed
Medical Systems, Auckland, New Zealand). Heifers
that had a corpus luteum were returned to the herd
without treatment, but heifers that were corpus luteum
(CL)-negative (n = 4 NEG heifers) had a reproductive
treatment to stimulate ovulation. Animals received an
intravaginal P4-releasing device (CIDR, Zoetis New
Zealand Limited, Auckland, New Zealand) from 0 to 7
d (insertion = d 0), gonadorelin (Ovurelin 100 mg i.m.;
Bomac Laboratories Ltd., Auckland, New Zealand) on
0 d, and 500 mg of cloprostenol i.m. on 7 d (Ovuprost,
Bayer Animal Health NZ, Auckland, New Zealand).
Heifer Breeding
We maintained the heifers in their 4 grazing herds
throughout the 98-d breeding season (starting Octo-
ber 4, 2016). Thirty-five 15-mo-old Jersey bulls were
commingled with each of the grazing herds at a ra-
tio of 1 bull per 20 heifers (6–8 bulls per group) with
the remaining bulls held in reserve to be rotated on
a regular basis. The Jersey bulls were sourced from a
single supplier, were health and fertility tested (before
the breeding season), vaccinated for leptospirosis and
bovine viral diarrhea, and had BCS of ≥4.5 (scale of
1–10; Roche et al., 2004) 42 d before the start of the
breeding season.
Estrus Events and Estrus Rate
Estrus events were monitored using the SCR Hea-
time HR system (SCR Engineers Ltd., Netanya, Israel),
which included the collar-mounted Heatime sensor at-
tached to the upper left side of a collar worn at the
cranial part of each heifer’s neck at approximately 213
d before the start of the breeding season (200 d, SD
= 11.2 d, range 154–235 d). The Heatime sensors col-
lected both activity and rumination data (via micro-
phone) and sent data wirelessly every 2 h to a receiving
unit connected to a base computer (Burfeind et al.,
2011; Silper et al., 2015a,b). As the heifers did not visit
a central yarding point on a daily basis, data collec-
tion occurred via 9 receiver stations (routers; including
WIFI nodes), and 2 repeater units (with solar panels)
were deployed at high points close to water troughs
around the grazing property to allow for continuous
data transmission to the base computer. Each heifer’s
activity and rumination data were translated into an
index value (0–100) that represented weighted SD from
its own basal activity. A system heat was logged when
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
the threshold was reached for an episode of high activ-
ity, using the manufacturer’s setting (SCR Engineers
Ltd.).
We calculated the proportion of heifers with a SCR
system heat (SH) alert during the first 21 and 42 d of
the seasonal breeding period (SH21, SH42), as well
as the interval from the start of breeding to the first
SH alert.
Pregnancy Diagnoses, Pregnancy Rates, and Losses
Fetal aging was undertaken at 3 time points to enable
accurate pregnancy diagnosis and identify early embryo
losses. All heifers were examined 49 to 51 d after the
start of the breeding season. Nonpregnant heifers, or
those detected with a pregnancy less than 30 d old,
were enrolled for a second pregnancy diagnosis at 79 d
after the start of the breeding season. Confirmation of
pregnancy included identification of heartbeat to indi-
cate the presence of a viable fetus. The final pregnancy
diagnosis included all heifers and was undertaken 44 d
after bulls were removed. The method involved tran-
srectal ultrasonography using a 5- to 15-MHz probe
(SonoScape S6V, Euromed Medical Systems, Auckland,
New Zealand) or Esi-Scan using a 3- to 7-MHz probe
(BCF Technologies, Auckland, New Zealand).
Pregnancy rates were defined as the proportion of
heifers diagnosed pregnant by 21 (PR21), 42 (PR42),
63 (PR63), and 98 d (PR98) relative to the start of
the breeding season that were viable at the pregnancy
test (i.e., pregnancy losses are not included in the
pregnancy rate estimates). Pregnancy loss was defined
as a heifer that was pregnant at the first or second
pregnancy test, but was not pregnant or pregnant with
a younger fetus, at the final pregnancy diagnosis.
Statistical Analysis
We undertook the analyses using SAS/STAT 15.1
(SAS Institute Inc., 2018). Body weight was analyzed
as repeated measurements using random coefficient
model with fertility group (POS, NEG), age in days
up to third-degree polynomial, and their interactions
included as fixed effects, and cow, intercept, and day
included as random effects. The random coefficients
were specified to have bivariate normal distribution
(type = un), whereas ADG, stature, and BCS were
subjected to repeated measures ANOVA using mixed
models approach (Proc Mixed). Fertility group (POS,
NEG), month, and their interaction were included as
fixed effects, and cow, sire, grazing herds, and original
herd were included as random effects. The covariance
patterns model was autoregressive heterogeneous [type
= arh(1)] to account for increasing variances within
and decreasing correlations between measures with
increasing age. Results from the repeated measures are
presented as adjusted means with standard errors of
the difference.
Cox proportional hazard models (Proc PHReg) with
censoring variables were used to analyze age, BW, and
percentage of mature BW at puberty, as well as time
from the start of breeding to first SCR SH alert and
time to conception. The models included fertility group
(POS, NEG) as fixed effect and sire as random effect.
Date of birth (number of days after June 1, 2015) was
included as covariate for the analyses of puberty. For
time to first breeding and conception, covariates were
age at puberty or day of puberty relative to start of
breeding and BW at puberty (due to autocorrelation,
only 1 age variable was included at any single time).
Puberty observations were censored for those cows that
had not reached the puberty threshold by the end of
progesterone sampling and were allocated a puberty
date of +30 d relative to the start of seasonal breeding.
Observations for time from start of breeding to first
SCR SH and time to conception were censored if the
animal had not been bred or had not conceived by the
end of seasonal breeding, and were allocated a time of
+104 d relative to the start of breeding. Time to events
are presented as survival curves with 95% confidence
interval (CI), median, and hazard ratio (HR) with
95% CI. Hazard ratios for covariates were assessed as
offsets from the mean and expressed in units of 10 (d).
Probability of heifers reaching puberty by breeding
start date and reproductive variables were analyzed
using binary logistic regression (Proc GLIMMIX). Re-
productive parameters included SH21 and SH42, being
pregnant after 21, 42, 63, 84, and 98 d relative to the
start of breeding (PR21, PR42, PR63, PR84, PR98),
losing a pregnancy, and being pregnant after pregnancy
loss. The models included fertility group (POS, NEG)
as fixed effect, and sire as random effect. Date of birth
(number of days after June 1, 2015) were included as
covariate for the analysis of puberty. For reproductive
parameters, covariates used were age at puberty or day
of puberty relative to start of breeding and BW at pu-
berty (due to autocorrelation, only 1 age variable was
included at any single time). Results of event ratios are
presented as adjusted mean percentages and absolute
counts for POS and NEG fertility group, and relative
risk (RR) with 95% CI for POS versus NEG fertility.
Hazard ratios for continuous covariates with 95% CI
were assessed as offsets from the mean and expressed
in units of 10 (d).
Analysis of animal losses was undertaken with
Fisher’s exact 2 × 2 test. Three analyses were under-
taken: (1) losses associated with parentage errors as a
proportion of the total calves collected, (2) losses due
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
to health (unsound + deaths + euthanized or culled) as
a proportion of the total calves collected, and (3) total
losses (failed parentage + unsound + death + eutha-
nized or culled + not pregnant) as a proportion of the
total calves collected. Descriptive data of the categories
(failed parentage, unsound, deaths and euthanized or
culled) and the subcategories within each category
are presented in Supplemental Table S2 (https: / / data
.mendeley .com/ datasets/ 343t97cpdr/ 1).
RESULTS
The ADG, BW, Stature, and BCS
A significant fertility by month interaction for ADG
(P = 0.003; Table 2) was evident. The interaction be-
tween FertBV and time was such that NEG FertBV
heifers had a greater ADG between 9 to 12 mo and
12 to 15 mo of age (P < 0.01; Table 2), with an ADG
advantage of 0.02 kg/d between 15 and 18 mo (P =
0.056). Average daily gain was least from 4 d to 6 mo
of age and 9 to 12 mo of age (0.58 and 0.64 kg/d),
periods that align with late winter to early spring, and
the following autumn to winter, respectively. There was
an increase in ADG between 12 and 15 mo of age (0.88
and 0.92 kg/d), corresponding to the next spring to
early summer.
Significant fertility by day interaction for BW was
evident (P < 0.001; Figure 1). This interaction was
such that the heifers with NEG genetic merit for fertil-
ity traits were increasingly heavier as the heifers aged,
such that the NEG heifers were 8 kg heavier on average
by 21 mo of age (NEG = 470 kg, POS = 462 kg, stan-
dard error of the difference = 2.9 kg; Figure 1). There
was no effect of FertBV nor interactions with age (mo)
on heifer girth, length, height, nor BCS (Table 2).
Puberty and Reproductive Parameters
Heifers with POS genetic merit for fertility traits
reached puberty earlier and at a lighter BW and lesser
percentage of mature BW. The median age, BW, and
percentage of mature BW for the POS heifers was 358
d, 274 kg, and 51%, respectively, and the NEG heifers
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Table 2. The ADG, girth, length, height, and BCS for the heifers with positive or negative genetic merit for fertility traits; data are presented
as adjusted means and the standard error of the difference (SED)
Variable Age
Genetic merit for fertility traits
Model P-value1
Positive Negative SED P-values2Fertility Mo Fert × Mo
ADG (kg/d) Mean 0.74 0.75 0.008 0.147 0.147 <0.001 0.003
4 d–3 mo 0.63 0.64 0.010 0.555 — — —
3–6 mo 0.63 0.62 0.011 0.466 — — —
6–9 mo 0.80 0.79 0.013 0.658 — — —
9–12 mo 0.58 0.62 0.013 0.002 — — —
12–15 mo 0.88 0.92 0.013 0.003 — — —
15–18 mo 0.88 0.91 0.019 0.056 — — —
18–21 mo 0.76 0.73 0.029 0.458 — — —
Girth (cm) Mean 144 145 0.4 0.377 0.377 <0.001 0.260
6 mo 124 124 0.5 0.908 — — —
9 mo 139 139 0.5 0.875 — — —
12 mo 150 151 0.5 0.323 — — —
15 mo 165 166 0.4 0.053 — — —
Length (cm) Mean 104 103 0.4 0.230 0.230 <0.001 0.401
6 mo 91 90 0.5 0.095 — — —
9 mo 101 100 0.4 0.114 — — —
12 mo 107 106 0.4 0.533 — — —
15 mo 117 117 0.5 0.933 — — —
Height (cm) Mean 109 109 0.4 0.893 0.893 <0.001 0.561
6 mo 97 97 0.5 0.921 — — —
9 mo 106 106 0.5 0.963 — — —
12 mo 113 113 0.5 0.834 — — —
15 mo 119 120 0.5 0.514 — — —
BCS3Mean 5.1 5.1 0.02 0.926 0.926 <0.001 0.658
6 mo 4.7 4.8 0.03 0.629 — — —
9 mo 5.0 4.9 0.03 0.667 — — —
12 mo 5.2 5.2 0.04 0.562 — — —
15 mo 5.4 5.4 0.03 0.742 — — —
1Repeated measures analyses, P-values for genetic merit for fertility traits (fertility), linear age in months (mo), and their interaction (Fert ×
Mo).
2P-values for genetic merit for fertility traits (positive vs. negative) at each age (mo).
3BCS scored on a 1–10 scale (Roche et al., 2004, 2007)
Journal of Dairy Science Vol. 104 No. 3, 2021
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Figure 1. Average BW of heifers with positive (solid line) or negative (dashed line) genetic merit for fertility traits from 8 to 644 d of age.
Data represent the estimated means and 95% CI (dotted lines) for each group. The standard error of the differences (SED) are not included due
to scale (SED range: positive, 0.89–2.86 kg; negative, 0.93–3.02 kg). Fertility, P < 0.001; quadratic fertility × day, P < 0.001. Data are arbitrarily
grouped as follows: (A) d 5–75, (B) 76–180 d, (C) 181–270 d, (D) 271–360 d, (E) 361–450 d, (F) 451–540 d, and (G) 541–644 d, respectively.
Journal of Dairy Science Vol. 104 No. 3, 2021
were 385 d, 294 kg, and 55%, respectively (Figure 2).
The HR for reaching puberty was greater in POS than
NEG heifers for age at puberty (HR = 1.98, 95% CI =
1.45–2.70, P < 0.001), BW at puberty (HR = 2.37, 95%
CI = 1.71–3.29, P < 0.001), and percentage of mature
BW at puberty (HR = 2.25, 95% CI = 1.68–3.01, P <
0.001). Figure 2 depicts the proportion of heifers reach-
ing puberty with increasing age, BW, or percentage
mature BW.
At the start of the seasonal breeding period, 94 ±
1.6% of the POS and 82 ± 3.2% of the NEG heifers
had reached puberty (P < 0.001). Indeed, POS genetic
merit for fertility traits, relative to NEG, had a 14%
greater chance of reaching puberty before the start of
the seasonal breeding period (RR = 1.14, 95% = CI 1.0
–1.18). The chance of reaching puberty at the start of
breeding was dependent on date of birth, such that for
every 10 d born later, the chance of reaching puberty
decreased by 9% (HR = 0.90, 95% CI = 0.86–0.94, P
< 0.001).
There was no difference in the proportion of heifers
that had a recorded SH during the first 21 or 42 d of
breeding season (SH21 or SH42; Table 3). The median
time from planned start of breeding to first SH was 11.3
d for the POS and 12.1 d for the NEG FertBV heifers.
There were no effects of age at puberty, BW at pu-
berty, nor reaching puberty before start of breeding on
whether the heifer had a recorded SH alert (Table 3).
The difference in heifer pregnancy rate at PR21 and
PR42 between the POS and NEG group was 12.6 and
8.6 percentage points in favor of the POS heifers, re-
spectively (P = 0.025, and P = 0.032; Table 3). As
breeding progressed, the difference between the POS
and NEG heifers reduced to 5 percentage points, with a
difference of 4.2 percentage points at the end of breed-
ing (P = 0.039; Table 3). POS heifers were 20% and
11% more likely to be pregnant after 21 d and 42 d
of breeding than NEG heifers (RR = 1.20, 95% CI =
1.03–1.34, P = 0.025; RR = 1.11, 95% CI = 1.01–1.16,
P = 0.032). With few pregnancy losses, there was no
difference between the POS and NEG heifers. There
were no effects of age and BW at puberty or puberty
relative to breeding on the PR parameters.
The POS FertBV heifers conceived earlier than the
NEG heifers (Figure 3). The heifers with POS genetic
merit for fertility traits conceived 3.6 d earlier (median
13.0 vs. 16.6 d, P = 0.001). At any given time during
the breeding period, 40% more POS heifers conceived
(P = 0.001; Table 4) compared with the NEG fertil-
ity heifers. Neither age, BW at puberty, nor puberty
expressed as days relative to the start of breeding were
associated with time conception (Table 4).
Animal Losses
The sources of animal losses between the time calves
were collected (~9 d of age) and final pregnancy diag-
nosis are described in Table 5. There were no differ-
ences in the proportion of heifers with POS or NEG
genetic merit for fertility traits that failed parentage
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Figure 2. Survival estimations from the Cox proportional hazard
model of (A) age (d, P < 0.001), (B) BW (kg, P < 0.001), and (C) per-
centage of estimated mature BW (%) at puberty (P < 0.001) of heifers
with positive (solid line) and negative (dashed line) genetic merit for
fertility traits. Dotted lines represent the 95% CI.
Journal of Dairy Science Vol. 104 No. 3, 2021
testing (P = 0.54). Significantly more heifers with NEG
genetic merit for fertility traits were removed due to ill
health compared with the POS group (POS, 14/324;
NEG, 27/316; P = 0.034). For total removals, fewer
heifers with POS genetic merit for fertility traits were
removed (55/324) compared with the NEG fertility
group (83/316; P < 0.01).
DISCUSSION
The earlier onset of puberty (younger and lighter) and
greater pregnancy rate in heifers with a POS compared
with NEG FertBV support our hypothesis that heifers
divergent in genetic merit for fertility traits differ in
their reproductive performance. To our knowledge, this
is the first reported example in which direct selection
for genetic merit for fertility traits, estimated using
reproductive traits from lactating cows, has resulted
in an earlier onset of heifer puberty. The effects of the
FertBV on heifer reproductive phenotypes reported
here align with recent findings of the effect of genetic
selection on detailed reproductive phenotypes (Cum-
mins et al., 2012a,b; Veronese et al., 2019a,b).
This indirect selection for earlier puberty occurred
even though the FertBV consisted of 6 binomial repro-
ductive traits related to recalving rates during lacta-
tions 2, 3, and 4, and breeding rates during the first 3
wk of seasonal breeding collected during lactations 1, 2,
and 3 (DairyNZ, 2018). Previous studies have reported
an indirect effect on puberty when selecting for pro-
ductivity traits. In a study evaluating the effects of 20
yr of genetic improvement in New Zealand dairy cows,
greater overall genetic merit led to heifers reaching
puberty later. It was identified that New Zealand heif-
ers representing the genetic potential from the 1970s
reached puberty earlier compared with New Zealand
heifers with genetics from the 1990s (Macdonald et al.,
2007). In the same study, both groups of New Zea-
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Table 3. Effect of genetic merit for fertility traits (positive or negative) and age at puberty (Age Pub), BW at puberty (BW Pub), and time of
puberty relative to the start of breeding (Day rel BS) on heifer reproductive parameters; data are presented as adjusted group mean proportions
(counts), and relative risk (RR) with 95% CI for the effect fertility, and group mean estimates with SEM for potential confounders
Variable
Genetic merit for fertility traits
Confounder
Positive Negative P-value RR Confounder RR (95% CI) P-value
Total heifers1 (n) 275 248
PR21274.9% (205) 62.3% (163) 0.025 1.20 (1.03–1.34) Age Pub 1.01 (0.99–1.03) 0.571
BW Pub 1.02 (0.99–1.05) 0.246
Days rel BS 1.00 (0.99–1.02) 0.898
PR42290.1% (247) 81.5% (204) 0.032 1.11 (1.01–1.16) Age Pub 1.00 (0.98–1.01) 0.686
BW Pub 1.01 (0.99–1.02) 0.359
Days rel BS 0.99 (0.98–1.00) 0.204
PR63293.4% (256) 87.9% (219) 0.073 1.06 (0.99–1.10) Age Pub 1.00 (0.99–1.01) 0.739
BW Pub 1.01 (0.99–1.02) 0.445
Days rel BS 0.99 (0.98–1.00) 0.162
PR98296.5% (264) 91.2% (227) 0.033 1.06 (1.01–1.08) Age Pub 1.01 (1.00–1.01) 0.124
BW Pub 1.00 (0.98–1.01) 0.508
Days rel BS 1.00 (1.00–1.01) 0.473
FinPR298.0% (269) 93.8% (232) 0.039 1.04 (1.00–1.06) Age Pub 1.00 (1.00–1.01) 0.158
BW Pub 0.99 (0.98–1.00) 0.210
Days rel BS 1.00 (1.00–1.01) 0.716
Pregnancy loss31.9% (5) 3.3% (9) 0.523 0.57 (0.00–29.9) Age Pub 0.94 (0.29–2.91) 0.615
BW Pub 1.13 (0.24–4.96) 0.497
Days rel BS 1.01 (0.38–2.65) 0.909
Pregnant after loss 72.2% (3) 16.9% (2) 0.326 4.28 (0.00–64.9) Age Pub 0.00 (0.00–1.43) 0.872
BW Pub 0.34 (0.00–4.8) 0.641
Days rel BS 0.00 (0.00–1.25) 0.329
SH21478.0% (215) 81.2% (201) 0.552 0.96 (0.18–1.21) Age Pub 1.00 (0.89–1.08) 0.921
BW Pub 1.00 (0.84–1.11) 0.823
Days rel BS 1.00 (0.90–1.07) 0.729
SH42482.5% (228) 85.6% (211) 0.541 0.96 (0.17–1.16) Age Pub 1.00 (0.91–1.07) 0.634
BW Pub 0.99 (0.84–1.08) 0.505
Days rel BS 1.00 (0.92–1.06) 0.893
1One heifer (negative) was excluded from analysis, as she was euthanized before final pregnancy diagnosis.
2Pregnancy rates are defined as the proportion of heifers identified as pregnant by d 21 (PR21), 42 (PR42), 63 (PR63), 98 (PR98, end of the sea-
sonal breeding) of the seasonal breeding period, where the pregnancy was still viable at the final pregnancy test 44–45 d after bulls were removed.
3Losses between confirmed pregnant and the pregnancy diagnoses on February 16–17 2017 (30–120 d of gestation, approximately).
4SCR system heat (the automated monitoring of estrus events using the SCR Heatime HR system; SCR Engineers Ltd., Netanya, Israel) alert
during the 21 (SH21) and 42 (SH42) d of the breeding season.
Journal of Dairy Science Vol. 104 No. 3, 2021
land heifers (1970s and 1990s) reached puberty earlier
than heifers with 1990s North American genetics and
at a lighter percentage of mature BW (McNaughton
et al., 2002; Macdonald et al., 2007). In other studies
(Garcia-Muniz, 1998) identified that HF heifers from
lines with greater mature BW (with high proportion
of North American ancestry) and larger stature were
older and heavier when they reached puberty when
compared with heifers with low genetic merit for ma-
ture BW (with predominantly New Zealand ancestry)
or smaller in stature. Hence, the onset of puberty can
be influenced by numerous factors, such as ancestry,
mature BW, and fertility traits from lactating cows.
To better understand these factors, data not biased by
our study design (selection for POS and NEG FertBV)
is needed. Therefore, focus should be on generating an
unbiased data set to robustly determine correlations
among ancestry, BW, fertility traits, and management
factors on the onset of puberty.
The extent of the differences in the age and BW at
puberty between the heifers with POS and NEG FertBV
were unexpected. The difference in the onset of puberty
reported here is comparable with changes in the onset
of puberty reported previously when undertaking single
trait section. For example, in a long-term study focused
on direct genetic selection for earlier puberty, over 7-yr
of single selection puberty was shifted by 62 d (Morris
and Amyes, 2005; Morris et al., 2011). This large shift
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Figure 3. Survival estimations from the Cox proportional hazard model of time to conception (d, P = 0.001) of heifers with positive (solid
line) and negative (dashed line) genetic merit for fertility traits. Dotted lines represent the 95% CI.
Table 4. Effect of genetic merit for fertility traits (positive or negative) and potential confounders on heifer reproductive parameters; data are
presented as median time from start of breeding to SCR system heat1 and to conception, and hazard ratio (HR) with 95% CI for the effect
fertility and per 10 d for potential confounders
Variable
Positive
(median)
Negative
(median) P-value
HR
(95% CI) Effect2HR per 10 d
(95% CI) P-value
Time to system heat (d) 11.3 12.1 0.305 1.11 (0.91–1.36) Age Pub 1.00 (0.99–1.00) 0.798
BW Pub 0.99 (0.99–1.01) 0.526
Days rel BS 0.99 (0.99–1.00) 0.979
Time to conception (d) 13.0 16.6 <0.001 1.40 (1.15–1.72) Age Pub 1.00 (0.99–1.00) 0.663
BW Pub 1.00 (0.99–1.01) 0.376
Days rel BS 0.99 (0.99–1.00) 0.246
1SCR system heat refer to the automated monitoring of estrus events using the SCR Heatime HR system (SCR Engineers Ltd., Netanya, Israel),
which included the collar-mounted Heatime sensor. A system heat was logged when the threshold was reached for an episode of high activity,
using the manufacturer’s setting.
2Age Pub = age at puberty; BW Pub = body weight at puberty; Days rel BS = days relative to the start of the breeding season (BS).
Journal of Dairy Science Vol. 104 No. 3, 2021
in the onset of puberty was possible because of the
single trait selection approach as well as the heritabil-
ity of puberty traits. The heritability of puberty has
a large reported range from 0.10 to 0.67 in beef and
dairy heifers. This range in heritability reflects both
the measures used and the study size (Martin et al.,
1992; Morris and Hickey, 2004). From our results, the
difference in the age at puberty between the POS and
NEG FertBV heifers suggests early onset of puberty is
correlated with the FertBV. The mechanisms altered to
result in such large differences remain to be elucidated.
In the current study, there was no difference in stat-
ure nor BCS, even though NEG FertBV heifers had a
small numeric advantage with BW and ADG between
9 to 21 mo of age. As previously discussed, mature BW
and genetic ancestry can affect the onset of puberty,
and the breeding approach resulted in the NEG FertBV
heifers having 4 kg greater BW BV and 6% greater
North American ancestry (Table 1). This small effect
is suggestive of the NEG FertBV having greater size,
although there was no difference in stature of the heif-
ers when evaluated at 6, 9, 12, and 15 mo of age. What
proportion of the difference in puberty is explained by
the difference in BW BV and ancestry remains to be
determined.
Management of heifers has a significant effect on when
heifers reach puberty. Previous studies have identified
that the age at puberty is inversely related to ADG
or nutrition levels, such that heifers with low ADG
are older at puberty (Patterson et al., 1992; Schillo et
al., 1992; Macdonald et al., 2005). Yet, we observed
that the NEG FertBV heifers were heavier and had a
greater ADG after they reached 9 mo of age, and the
NEG FertBV had a greater ADG up to 15 mo of age
(approximately 0.04 kg/d, which is equivalent to 1.2
kg BW over 30 d). Based on the information available,
we propose that in this study, ADG and nutrition were
not the main contributors to the difference in the onset
of puberty reported. The role of body composition and
stature at maturity is to be determined.
Industry recommendations identify that the average
heifer reaches puberty between 43 and 47% of mature
BW (DairyNZ, 2016b). These recommendations align
with the range reported by McNaughton et al. (2002)
of the 2 New Zealand strains (1990s and 1970s) at 43%
of mature BW, and the North American strain reaching
puberty at 47% of mature BW. Our results identified
that the heifers reached puberty at 51% and 55% of
mature BW. It remains to be seen whether the indus-
try expectations that the average heifer on commercial
farms reaches puberty at 43 to 47% of mature BW
continues to be appropriate. To ensure industry recom-
mendations are robust, estimates of BW at puberty
from commercial herds should be evaluated.
If ADG and nutrition are not the key factors control-
ling the onset of puberty as previously reported (Mac-
donald et al., 2005; Patterson et al., 1992; Schillo et al.,
1992), the underlying mechanisms controlling the differ-
ence in puberty in this study remain to be determined.
As the current study selected for extremes in genetic
merit for fertility traits, it is plausible that inherent
difference in the hypothalamus and pituitary signals
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Table 5. Heifer losses between 9 d of age (at collection) and the final pregnancy diagnosis (>18 mo of age) for the 2 lines of positive (POS) or
negative (NEG) genetic merit for fertility traits
Variable
Genetic merit for fertility traits
Total P-valuePOS NEG
Total collected (n) 324 316 640
Failed parentage verification to sire or maternal grandsire (n) 35 40 75
(%)1,2,3 (10.8) (12.7) (11.7)
Unsound3,4 (conformation/freemartin; n) 2 6 8
(%)1(0.6) (1.9) (1.3)
Deaths3,4 (n) 6 14 20
(%)1(1.9) (4.4) (3.1)
Euthanized or culled3,4 (n) 6 7 13
(%)1(1.9) (2.2) (2.0)
Not pregnant3 (n) 6 16 22
(%)1(2.2) (6.4) (3.4)
Heifers remaining (n; May 2017) 269 233 502 <0.01
(%)1(83.0) (73.7) (78.4)
1Percentages of those heifers collected.
2Failed parentage Fishers exact 2 × 2 test: POS, 35 from 324; NEG, 40 from 316; P = 0.54.
34All losses due to parentage failure, health (unsound, deaths, euthanized or culled) and not pregnant. Fishers exact 2 × 2 test: POS, 55 from
324; NEG, 83 from 316; P < 0.01.
4Losses due to health (unsound, deaths, euthanized or culled) after calves with failed parentage are removed. Fishers exact 2 × 2 test: POS, 14
from 289; NEG, 27 from 270; P = 0.035.
Journal of Dairy Science Vol. 104 No. 3, 2021
determine the timing of puberty. A deeper knowledge
of whether the biological mechanisms that control pu-
berty differ between the POS and NEG FertBV lines
may support the discovery of new candidate traits that
benefit cow reproductive performance.
The earlier puberty in the heifers with POS genetic
merit for fertility traits meant that these heifers had 1
more estrus event, on average, before the start of breed-
ing. This can provide significant effects on pregnancy
outcomes, as heifers bred on the second or third estrus
have 36% greater conception and 20% greater pregnan-
cy rates compared with those bred on the first estrus
(Byerley et al., 1987; Perry et al., 1991). Our finding
aligns with that of Funston et al. (2012), who reported
that overall pregnancy rates in heifers was directly in-
fluenced by the proportion of heifers showing estrus
before the beginning of the breeding season. Future
solutions that aim to optimize reproductive outcomes
for seasonally bred heifers should be cognizant of the
gains that could be achieved if heifers are postpubertal
(second or third estrus) early in the breeding season.
In the current study, the focus was on the benefits
within the current breeding season. However, long-
term benefits have also been reported. Heifers that are
well grown, and heifers that get pregnant early in the
breeding season, calve earlier and have improved life-
time production and reproduction (Pryce et al., 2007;
Wathes et al., 2014; Dennis et al., 2018; Handcock et
al., 2020). The extent and consistency of benefits un-
der commercial conditions requires a larger data set to
quantify the benefits under commercial environments.
Additional value may be generated by understanding
these relationships across different farm systems (sea-
sonal twice a day milking, once a day milking, split
calving in spring and autumn, year-round calving).
The breeding priorities of many countries are focused
on breeding cows most suited for that specific dairying
industry (dairy system), with increasing importance
on a balance between productivity, profitability, and
robustness. This focus has put more emphasis on
breeding for traits associated with cow reproductive
performance and health (Miglior et al., 2005; Cole
and VanRaden, 2018). Yet, few breeding approaches
include heifer traits, and none include heifer puberty.
Three points that make age at puberty an attractive
candidate trait to consider in selection indices are as
follows: (1) the heritability is better than that reported
for traditional traits in use currently for estimating ge-
netic merit for fertility, (2) puberty occurs earlier in life
than the current (lactational) traits used, and (3) there
are benefits in heifers reaching puberty and conceiving
early with respect to their longevity in the herd. We
believe that generating data sets that support robust
evaluation of genetic and phenotypic correlations be-
tween puberty, heifer conception and pregnancy rate,
and traits currently in the animal evaluation models is
the next step to progressing this area. The difficulties
will be associated with achieving appropriate record-
ing (widespread or targeted approach), and acceptance
that these phenotypes may be from a limited number of
heifers (reference population). Puberty data on a large
scale could be estimated using plasma progesterone or
automated systems to capture puberty (pedometer or
activity collars). There are, however, trade-offs that
will need to be accepted including frequency of data,
bias data sets, accuracy, and volume of data that can
be collected.
CONCLUSIONS
In the current study, we demonstrated that select-
ing for extreme positive POS (+5) genetic for fertil-
ity based on the New Zealand FertBV estimated from
predictor traits collected during lactations 1 to 4 will
produce heifers that reach puberty earlier, with greater
pregnancy rates during their first breeding period. This
effect is independent of heifer growth rates. Our results
indicated that heifer puberty and pregnancy rate are
potential earlier predictor traits than the cow fertility
traits used currently. Furthermore, understanding how
selection for genetic merit for fertility traits has altered
the physiological and genetic mechanisms controlling
puberty may provide additional early indicators for
subsequent cow fertility.
ACKNOWLEDGMENTS
This project was funded by a partnership
(DRCX1302) between the New Zealand Ministry of
Business, Innovation and Employment (Wellington,
New Zealand) and New Zealand dairy farmers through
DairyNZ Inc. (Hamilton, New Zealand) and includes
AgResearch SIFF funding (Hamilton, New Zealand).
A large contribution to this study also included in-kind
support from LIC (Hamilton, New Zealand) and CRV-
Ambreed (Hamilton, New Zealand) during the plan-
ning phase. Jack Hooper, Anna Burke, Katie Eketone,
and other members of the LIC team are gratefully ac-
knowledged for their expertise during the development
phase and successfully managing the contract mating,
communications with farmers, and calf collection. We
also acknowledge further contributions of LIC by pro-
viding data for the establishment of this research herd.
Ben Fisher, Kelly Collier, Stuart Morgan, and other
members of the DairyNZ technical and farm staff are
gratefully acknowledged for successfully executing the
challenges associated with calf collection, the measures
and samples collected, and data collation. Plasma sam-
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
ples were analyzed for progesterone by Angela Sheahan
, and Barbara Dow and Barbara Kuhn-Sherlock (all
of DairyNZ) supported the statistical analyses for this
study. The input from the calf rearer, the grazier, and
their respective staff is gratefully acknowledged. We
acknowledge the support of Claire Phyn (DairyNZ)
and Eric Hillerton in reviewing this manuscript pre-
submission. This study could not have occurred with-
out the participation of the New Zealand dairy farmers
who supplied the dams and calves for this project. The
authors have not stated any conflicts of interest.
REFERENCES
Bamber, R. L., G. E. Shook, M. C. Wiltbank, J. E. Santos, and P. M.
Fricke. 2009. Genetic parameters for anovulation and pregnancy
loss in dairy cattle. J. Dairy Sci. 92:5739–5753. https: / / doi .org/ 10
.3168/ jds .2009 -2226.
Berry, D. P., E. Wall, and J. E. Pryce. 2014. Genetics and genom-
ics of reproductive performance in dairy and beef cattle. Animal
8(Suppl. 1):105–121. https: / / doi .org/ 10 .1017/ S1751731114000743.
Bowley, F. E., R. E. Green, P. R. Amer, and S. Meier. 2015. Novel
approaches to genetic analysis of fertility traits in New Zealand
dairy cattle. J. Dairy Sci. 98:2005–2012. https: / / doi .org/ 10 .3168/
jds .2014 -8266.
Burfeind, O., K. Schirmann, M. A. von Keyserlingk, D. M. Veira, D.
M. Weary, and W. Heuwieser. 2011. Technical note: Evaluation of
a system for monitoring rumination in heifers and calves. J. Dairy
Sci. 94:426–430. https: / / doi .org/ 10 .3168/ jds .2010 -3239.
Byerley, D. J., R. B. Staigmiller, J. G. Berardinelli, and R. E. Short.
1987. Pregnancy rates of beef heifers bred either on puberal or
third estrus. J. Anim. Sci. 65:645–650. https: / / doi .org/ 10 .2527/
jas1987 .653645x.
Carthy, T. R., D. P. Berry, A. Fitzgerald, S. McParland, E. J. Williams,
S. T. Butler, A. R. Cromie, and D. Ryan. 2014. Risk factors associ-
ated with detailed reproductive phenotypes in dairy and beef cows.
Animal 8:695–703. https: / / doi .org/ 10 .1017/ S1751731114000354.
Cole, J. B., and P. M. VanRaden. 2018. Symposium review: Possibili-
ties in an age of genomics: The future of selection indices. J. Dairy
Sci. 101:3686–3701. https: / / doi .org/ 10 .3168/ jds .2017 -13335.
Cummins, S. B., P. Lonergan, A. C. Evans, D. P. Berry, R. D. Ev-
ans, and S. T. Butler. 2012a. Genetic merit for fertility traits in
Holstein cows: I. Production characteristics and reproductive ef-
ficiency in a pasture-based system. J. Dairy Sci. 95:1310–1322.
https: / / doi .org/ 10 .3168/ jds .2011 -4742.
Cummins, S. B., P. Lonergan, A. C. Evans, and S. T. Butler. 2012b.
Genetic merit for fertility traits in Holstein cows: II. Ovarian fol-
licular and corpus luteum dynamics, reproductive hormones, and
estrus behavior. J. Dairy Sci. 95:3698–3710. https: / / doi .org/ 10
.3168/ jds .2011 -4976.
Cummins, S. B., S. M. Waters, A. C. Evans, P. Lonergan, and S. T.
Butler. 2012c. Genetic merit for fertility traits in Holstein cows:
III. Hepatic expression of somatotropic axis genes during preg-
nancy and lactation. J. Dairy Sci. 95:3711–3721. https: / / doi .org/
10 .3168/ jds .2011 -4977.
DairyNZ. 2016a. Interpreting the info: Genetic base cow. Accessed Dec.
7, 2016. https: / / www .dairynz .co .nz/ animal/ animal -evaluation/
interpreting -the -info/ genetic -base -cow/ .
DairyNZ. 2016b. Liveweight Targets. Accessed Dec. 16, 2016. https: / /
www .dairynz .co .nz/ animal/ heifers/ liveweight -targets/ .
Dairy, N. Z. 2017. Interpreting the info: Economic values. Ac-
cessed Mar. 31. 2017. https: / / www .dairynz .co .nz/ animal/ animal
-evaluation/ interpreting -the -info/ economic -values/ .
DairyNZ. 2018. Description of National Genetic Evaluation Systems:
Female Fertility. Accessed May 4, 2018. https: / / www .dairynz .co
.nz/ animal/ animal -evaluation/ interpreting -the -info/ breeding
-values/ .
Dennis, N. A., K. Stachowicz, B. Visser, F. S. Hely, D. K. Berg, N. C.
Friggens, P. R. Amer, S. Meier, and C. R. Burke. 2018. Combin-
ing genetic and physiological data to identify predictors of lifetime
reproductive success and the effect of selection on these predictors
on underlying fertility traits. J. Dairy Sci. 101:3176–3192. https: /
/ doi .org/ 10 .3168/ jds .2017 -13355.
Egger-Danner, C., J. B. Cole, J. E. Pryce, N. Gengler, B. Heringstad,
A. Bradley, and K. F. Stock. 2015. Invited review: Overview of
new traits and phenotyping strategies in dairy cattle with a focus
on functional traits. Animal 9:191–207. https: / / doi .org/ 10 .1017/
S1751731114002614.
Fleming, R., D. B. Seifer, J. L. Frattarelli, and J. Ruman. 2015. As-
sessing ovarian response: Antral follicle count versus anti-Mulleri-
an hormone. Reprod. Biomed. Online 31:486–496. https: / / doi .org/
10 .1016/ j .rbmo .2015 .06 .015.
Funston, R. N., J. L. Martin, D. M. Larson, and A. J. Roberts. 2012.
Physiology and endocrinology symposium: Nutritional aspects of
developing replacement heifers. J. Anim. Sci. 90:1166–1171. https:
/ / doi .org/ 10 .2527/ jas .2011 -4569.
Garcia-Muniz, J. G. 1998. Studies of Holstein-Friesian cattle bred for
heavy or light mature live weight. PhD thesis. Institute of Veteri-
nary, Animal and Biomedical Sciences, Massey University, Palm-
erston North, New Zealand.
Handcock, R. C., N. Lopez-Villalobos, L. R. McNaughton, P. J. Back,
G. R. Edwards, and R. E. Hickson. 2020. Body weight of dairy
heifers is positively associated with reproduction and stayability. J.
Dairy Sci. 103:4466–4474. https: / / doi .org/ 10 .3168/ jds .2019 -17545.
Harris, B. L., J. E. Pryce, Z. Z. Xu, and W. A. Montgomerie. 2006.
Development of new fertility breeding values in the dairy industry.
Pages 107–112 in Proc. N.Z. Soc. Anim. Prod., Napier, New Zea-
land. N.Z. Soc. Anim. Prod., Hamilton, New Zealand.
Jenkins, G. M., P. Amer, K. Stachowicz, and S. Meier. 2016. Pheno-
typic associations between gestation length and production, fertil-
ity, survival, and calf traits. J. Dairy Sci. 99:418–426. https: / / doi
.org/ 10 .3168/ jds .2015 -9934.
Kinghorn, B. P. 2011. An algorithm for efficient constrained mate se-
lection. Genet. Sel. Evol. 43:4. https: / / doi .org/ 10 .1186/ 1297 -9686
-43 -4.
Lucy, M. C. 2019. Symposium review: Selection for fertility in the
modern dairy cow—Current status and future direction for genetic
selection. J. Dairy Sci. 102:3706–3721. https: / / doi .org/ 10 .3168/ jds
.2018 -15544.
Macdonald, K. A., L. R. McNaughton, G. A. Verkerk, J. W. Penno,
L. J. Burton, D. P. Berry, P. J. Gore, J. A. Lancaster, and C. W.
Holmes. 2007. A comparison of three strains of Holstein-Friesian
cows grazed on pasture: Growth, development, and puberty. J.
Dairy Sci. 90:3993–4003. https: / / doi .org/ 10 .3168/ jds .2007 -0119.
Macdonald, K. A., J. W. Penno, A. M. Bryant, and J. R. Roche. 2005.
Effect of feeding level pre- and post-puberty and body weight at
first calving on growth, milk production, and fertility in grazing
dairy cows. J. Dairy Sci. 88:3363–3375. https: / / doi .org/ 10 .3168/
jds .S0022 -0302(05)73020 -4.
Martin, L. C., J. S. Brinks, R. M. Bourdon, and L. V. Cundiff. 1992.
Genetic effects on beef heifer puberty and subsequent reproduc-
tion. J. Anim. Sci. 70:4006–4017. https: / / doi .org/ 10 .2527/ 1992
.70124006x.
McNaughton, L. R., S. R. Morgan, P. Gore, G. A. Verkerk, C. W.
Holmes, and T. J. Parkinson. 2002. Monitoring onset of puberty in
three genetic strains of Holstein-Friesian dairy cattle. Pages 30–33
in Proc. N.Z. Soc. Anim. Prod., Palmerston North, New Zealand.
N.Z. Soc. Anim. Prod., Hamilton, New Zealand.
Miglior, F. 2002. Overview of different breeding objectives in various
countries. Pages 1–9 in Proc. World Holst. Fries. Fed., Madison,
WI. World Holst. Fries. Fed, Killarney, Ireland.
Miglior, F., B. L. Muir, and B. J. Van Doormaal. 2005. Selection in-
dices in holstein cattle of various countries. J. Dairy Sci. 88:1255–
1263. https: / / doi .org/ 10 .3168/ jds .S0022 -0302(05)72792 -2.
Mion, B., M. R. Carvalho, and E.S. Ribeiro. 2019. Association be-
tween reproductive efficiency of heifers and their reproductive per-
formance as lactating cows. J. Dairy Sci. 102(Suppl. 1):359.
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
Journal of Dairy Science Vol. 104 No. 3, 2021
Moore, S. G., T. Fair, P. Lonergan, and S. T. Butler. 2014. Genetic
merit for fertility traits in Holstein cows: IV. Transition period,
uterine health, and resumption of cyclicity. J. Dairy Sci. 97:2740–
2752. https: / / doi .org/ 10 .3168/ jds .2013 -7278.
Morris, C. A., and N. C. Amyes. 2005. Response to selection for age
at puberty in an Angus herd. Pages 157–160 in Proc. Assoc. Adv.
Anim. Breed. Genet., Queensland, Australia. Assocc. Adv. Anim.
Breed. Genet., University of New England, Armidale, Australia.
Morris, C. A., N. C. Amyes, and S. M. Hickey. 2011. Respons-
es of prolactin and hair growth to selection for age at puberty
in Angus cattle. Animal 5:198–201. https: / / doi .org/ 10 .1017/
S1751731110001825.
Morris, C. A., and S. M. Hickey. 2004. Heritability of puberty traits
in dairy heifers in commercial herds. Pages 115–117 in Proc. N.Z.
Soc. Anim. Prod., Hamilton, New Zealand. Soc. Anim. Prod.,
Hamilton, New Zealand.
Morris, C. A., J. A. Wilson, G. L. Bennett, N. G. Cullen, S. M. Hickey,
and J. C. Hunter. 2000. Genetic parameters for growth, puberty,
and beef cow reproductive traits in a puberty selection experiment.
N. Z. J. Agric. Res. 43:83–91. https: / / doi .org/ 10 .1080/ 00288233
.2000 .9513411.
NZAgbiz. 2020. Feeding Guidelines. Accessed Aug. 15, 2020. https: / /
nzagbiz .co .nz/ nz/ en/ products/ calf/ ancalf .html.
Patterson, D. J., R. C. Perry, G. H. Kiracofe, R. A. Bellows, R. B.
Staigmiller, and L. R. Corah. 1992. Management considerations
in heifer development and puberty. J. Anim. Sci. 70:4018–4035.
https: / / doi .org/ 10 .2527/ 1992 .70124018x.
Perry, R. C., L. R. Corah, R. C. Cochran, J. R. Brethour, K. C. Olson,
and J. J. Higgins. 1991. Effects of hay quality, breed, and ovarian
development on onset of puberty and reproductive performance
of beef Heifers. J. Prod. Agric. 4:13–18. https: / / doi .org/ 10 .2134/
jpa1991 .0013.
Petersson, K. J., B. Berglund, E. Strandberg, H. Gustafsson, A. P.
Flint, J. A. Woolliams, and M. D. Royal. 2007. Genetic analysis of
postpartum measures of luteal activity in dairy cows. J. Dairy Sci.
90:427–434. https: / / doi .org/ 10 .3168/ jds .S0022 -0302(07)72644 -9.
Pryce, J. E., B. L. Harris, and L. R. McNaughton. 2007. The genetic
relationship between heifer and cow fertility. 67th Proc. New Zea-
land Society of Animal Production, Wanaka, New Zealand. N.Z.
Soc. Anim. Prod., Hamilton, New Zealand.
Pryce, J. E., R. Woolaston, D. P. Berry, E. Wall, M. Winters, R.
Butler, and M. Shaffer. 2014. World Trends in Dairy Cow Fertil-
ity. Pages 1–6 in the World Congress Genet. Appl. Livest. Prod.,
Vancouver, BC, Canada. Am. Soc. Anim. Sci., Champaign, IL.
Raheja, K. L., E. B. Burnside, and L. R. Schaeffer. 1989. Heifer fertil-
ity and its relationship with cow fertility and production traits in
Holstein dairy cattle. J. Dairy Sci. 72:2665–2669. https: / / doi .org/
10 .3168/ jds .S0022 -0302(89)79407 -8.
Roche, J. R., P. G. Dillon, C. R. Stockdale, L. H. Baumgard, and M.
J. VanBaale. 2004. Relationships among international body condi-
tion scoring systems. J. Dairy Sci. 87:3076–3079. https: / / doi .org/
10 .3168/ jds .S0022 -0302(04)73441 -4.
Roche, J. R., K. A. Macdonald, C. R. Burke, J. M. Lee, and D. P.
Berry. 2007. Associations among body condition score, body
weight, and reproductive performance in seasonal-calving dairy
cattle. J. Dairy Sci. 90:376–391. https: / / doi .org/ 10 .3168/ jds .S0022
-0302(07)72639 -5.
SAS Institute Inc. 2018. SAS/STAT, 15.1 User’s Guide. Cary, North
Carolina.
Schillo, K. K., J. B. Hall, and S. M. Hileman. 1992. Effects of nutrition
and season on the onset of puberty in the beef heifer. J. Anim. Sci.
70:3994–4005. https: / / doi .org/ 10 .2527/ 1992 .70123994x.
Silper, B. F., A. M. Madureira, M. Kaur, T. A. Burnett, and R. L.
Cerri. 2015a. Short communication: Comparison of estrus charac-
teristics in Holstein heifers by 2 activity monitoring systems. J.
Dairy Sci. 98:3158–3165. https: / / doi .org/ 10 .3168/ jds .2014 -9185.
Silper, B. F., L. Polsky, J. Luu, T. A. Burnett, J. Rushen, A. M.
de Passille, and R. L. Cerri. 2015b. Automated and visual mea-
surements of estrous behavior and their sources of variation in
Holstein heifers. II: Standing and lying patterns. Theriogenology
84:333–341. https: / / doi .org/ 10 .1016/ j .theriogenology .2014 .12 .030.
Tiezzi, F., C. Maltecca, A. Cecchinato, M. Penasa, and G. Bittante.
2012. Genetic parameters for fertility of dairy heifers and cows at
different parities and relationships with production traits in first
lactation. J. Dairy Sci. 95:7355–7362. https: / / doi .org/ 10 .3168/ jds
.2012 -5775.
Veronese, A., O. Marques, R. Moreira, A. L. Belli, R. S. Bisinotto,
T. R. Bilby, F. Penagaricano, and R. C. Chebel. 2019a. Genomic
merit for reproductive traits. I: Estrous characteristics and fertility
in Holstein heifers. J. Dairy Sci. 102:6624–6638. https: / / doi .org/ 10
.3168/ jds .2018 -15205.
Veronese, A., O. Marques, F. Penagaricano, R. S. Bisinotto, K. G.
Pohler, T. R. Bilby, and R. C. Chebel. 2019b. Genomic merit for
reproductive traits. II: Physiological responses of Holstein heif-
ers. J. Dairy Sci. 102:6639–6648. https: / / doi .org/ 10 .3168/ jds .2018
-15245.
Wathes, D. C., G. E. Pollott, K. F. Johnson, H. Richardson, and J. S.
Cooke. 2014. Heifer fertility and carry over consequences for life
time production in dairy and beef cattle. Animal 8(s1):91–104.
https: / / doi .org/ 10 .1017/ S1751731114000755.
ORCIDS
S. Meier https: / / orcid .org/ 0000 -0002 -4386 -7734
R. Handcock https: / / orcid .org/ 0000 -0001 -7017 -9948
P. R. Amer https: / / orcid .org/ 0000 -0002 -6428 -7165
J. R. Bryant https: / / orcid .org/ 0000 -0001 -8928 -0253
K. G. Dodds https: / / orcid .org/ 0000 -0002 -9347 -6379
J. R. Roche https: / / orcid .org/ 0000 -0002 -4165 -9253
C. R. Burke https: / / orcid .org/ 0000 -0003 -3868 -8675
Meier et al.: GENETIC MERIT FOR FERTILITY TRAITS ALTERS HEIFER REPRODUCTION
