Theriogenology 68 (2007) 682–686

www.theriojournal.com

Technical note
Biometry of frozen–thawed sperm from eight breeds
of Indian buffaloes (Bubalus bubalis)
R.A.K. Aggarwal a,*, S.P.S. Ahlawat, Y. Kumar a, P.S. Panwar a,
K. Singh b, M. Bhargava b
a
National Bureau of Animal Genetic Resources, Karnal 132001, India
b
CSK HP Agricultural University, Palampur 176062, India
Received 4 October 2006; accepted 20 March 2007

Abstract
Sperm morphometry, in combination with other objective traits, can be useful for developing a fertility index. The objective of
the present study was to measure various biometric end points of frozen–thawed sperm from eight breeds of Indian buffaloes
(Murrah, Surti, Tarai, Mehsana, Jaffrabadi, Bhadawari, Pandharpuri and Nili-Ravi). The sperm head of Pandharpuri buffaloes had
the greatest length (10.21 mm), width (6.05 mm), area (52.31 mm2) and perimeter (31.86 mm). The ratio of sperm width to length
was also greatest (0.61) in Pandharpuri as well as in two other breeds, viz. Nili-Ravi and Jaffrabadi. Murrah had the smallest sperm
head width (4.75 mm), area (41.65 mm2) and perimeter (29.17 mm), but its sperm tail was longest (57.02 mm), along with that of
Jaffrabadi buffaloes (56.96 mm). Based on mean values of sperm tail length, mid piece length and its width the eight buffalo breeds
were categorized into three, four and five groups, respectively. Multivariate analysis and clustering put six breeds (Surti, Tarai,
Mehsana, Jaffrabadi, Bhadawari and Nili-Ravi) in one cluster, whereas Murrah and Pandharpuri appeared as separate entities.
# 2007 Elsevier Inc. All rights reserved.

Keywords: Sperm; Cryopreservation; Frozen–thawed; Morphometry; Indian buffalo

1. Introduction explained by routine semen analysis [9,10]. Therefore,

Evaluation of sperm concentration and motility is
frequently used to assess semen quality, but provides
limited information regarding potential fertility of sires
[1–4]. Other criteria, including computerized analysis
of motility and acrosome integrity, have also been used
to assess semen quality; however, associations with
non-return rates of bulls are not high or even consistent
[5–8]. Non-return rates vary as much as 25% among
bulls from AI centres; these differences are not
* Corresponding author. Tel.: +91 184 2267918;
fax: +91 184 2267654.
E-mail addresses: rakaplp@yahoo.co.in, rakaplp@rediffmail.com
(R.A.K. Aggarwal).
sires with apparently normal semen that are subfertile
provide the impetus for further studies to elucidate other
markers of fertility. Sperm head morphology has been
suggested as an indicator of fertility [11]. Combining
head shape and sperm morphometry, as well as other
objective traits into an overall fertility index, could have
the potential to rank sires according to their fertilizing
capability.
India is a source of some of the best riverine breeds of
buffaloes. Murrah, Nili-Ravi and Surti enjoy a dominant
position among breeds noted for milk production.
Bhadawari is reported to have high milk fat and
Jaffrabadi is the heaviest of all the Indian buffalo breeds.
In the Indian subcontinent, there are 10 recognized
breeds [12] of buffaloes; they have important roles
as producers of milk, draught power, dung and other

0093-691X/$ – see front matter # 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.theriogenology.2007.03.028
 R.A.K. Aggarwal et al. / Theriogenology 68 (2007) 682–686
value-added products. However, one of the major
constraints for full exploitation of the productive
potential of buffalo has been its inherently low
reproductive efficiency, manifested as a low calving rate
and long generation interval [13].
Due to the potential role of sperm biometry in
assessment of sire fertility [11], the present study was
undertaken to evaluate sperm biometry in eight breeds
of Indian buffaloes.
2. Materials and methods
2.1. Samples
Sperm biometry was done on frozen–thawed semen
of eight breeds of Indian buffaloes (Murrah, Surti, Tarai,
Mehsana, Jaffrabadi, Bhadawari, Pandharpuri and Nili-
Ravi; five bulls for each breed). Straws of frozen semen,
cryopreserved according to a standard protocol [14],
were procured from several semen stations within India
(Murrah, NDRI, Karnal; Surti, LRS, Vallabhnagar;
Tarai, GBPU, Pantnagar; Mehsana, Dudhsagar Dairy,
Mehsana; Jaffrabadi, GAU, Junagarh; Bhadawari, Vety.
College, Mathura; Pandharpuri, RAU, Rahauri; Nili-
Ravi, LDB, Nabha).
2.2. Slide preparation and morphometrical
evaluation
Three straws of each bull (from different ejaculates)
were thawed in a water bath (37 8C for 30 s) and pooled.
A drop of the pooled semen was mixed with two or three
drops of eosin-nigrosin stain (Hi Media; 100 mg
eosin + 500 mg nigrosin dissolved in 10 mL of 2.9%
sodium citrate buffer, pH 6.9) and placed on a clean
glass slide at 37 8C. The mixture was mixed well and
then spread into a thin smear, which was then air dried
[15]. Sperm that had taken up the eosin were classified
as dead, whereas those without eosin were classified as
live. One hundred live sperm were used for sperm
biometry estimation from various fields on each of five
slides prepared for each bull (total of 500 sperm/bull).
Sperm were examined (1000 magnification) with a
Trinocular Research Phase Contrast Microscope (Motic
China Group, Ltd., Hong Kong) fitted with a high-
resolution digital CCD camera (still image
resolution = 768  576 pixels) and analyzed with
imaging software (Motic Images Plus 2.0; Motic China
Group, Ltd., Hong Kong). The system was first
calibrated with images of standard length for known
magnifications and measurement accuracy of 0.1 mm.
Various measurements of the sperm head (length, width,
683
perimeter and area) and sperm tail (length and width of
midpiece and length of tail) were recorded. The ratio of
width to length for sperm head was also calculated.
2.3. Statistical analysis
All breeds were categorized into groups using the
value of critical difference (CD) for each end point;
breeds within a group had similar values, but they were
significantly different from breeds of other groups. The
CD was estimated by first calculating the value of
standard error of difference (SED) through ANOVA,
performed using the PC-2 version of fixed model least
squares and maximum likelihood (LSML) computer
program [16], which included the effects of bulls and
breeds as sources of variation. The SED was then
multiplied with the t-value for P = 0.05 at error degrees
of freedom to arrive at the value of CD [17]. The
principal component analysis (PCA) for multivariate
analysis and clustering by the Tocher method were done
using the Windows 8.0 package developed by Indostat
Services, Hyderabad, India.
3. Results
3.1. Sperm head biometry
Based on average values of sperm head length and
the CD value, the eight breeds were categorized into
four groups (Table 1). The sperm head was longest
(10.21 mm) in Pandharpuri (Group ‘‘a’’) followed by
Mehsana, Surti and Bhadawari (Group ‘‘b’’) and
Jaffrabadi, Tarai and Murrah (Group ‘‘c’’); Nili-Ravi
(Group ‘‘d’’) had the shortest head length (9.33 mm).
Sperm width varied from 6.05 mm (Pandhapuri) to
4.75 mm (Murrah); there were five distinct groups.
Values for Tarai, Surti and Jaffrabadi (Group ‘‘b’’) were
similar but significantly higher than Nili-Ravi and
Mehsana (Group ‘‘c’’) and Bhadawari (Group ‘‘d’’).
Sperm head area and perimeter were greatest in
Pandharpuri and lowest in Murrah (Table 1). The sperm
head area in Tarai, Surti and Jaffrabadi was similar but
less than Pandharpuri. Bhadawari and Mehsana ranked
third in head area and had higher values than Nili-Ravi.
The sperm head perimeter in Tarai (31.48 mm) was less
than Pandharpuri, but greater than Surti and Jaffrabadi.
Values for this end point did not differ among
Bhadawari, Mehsana and Nili-Ravi.
The ratio of sperm head width to length varied from
0.50 in Murrah to 0.61 in Pandharpuri, enabling the
eight breeds to be allocated into four groups.
Pandharpuri, Nili-Ravi, Jaffrabadi had the highest
684 R.A.K. Aggarwal et al. / Theriogenology 68 (2007) 682–686

Table 1
Biometry of the sperm head in Indian buffaloes
Breed Maximum Maximum Area Perimeter Maximum width/
length (mm) width (mm) (mm2) (mm) maximum length
Bhadawari 9.82b* 5.31d 47.85c 30.79d 0.55c
(10.28)** (13.33) (8.65) (4.83) (22.09)
Mehsana 9.98b 5.42c 48.31c 30.65d 0.55c
(10.83) (14.1) (11.48) (6.32) (22.42)
Murrah 9.69c 4.75e 41.65e 29.17e 0.50d
(9.09) (13.38) (10.56) (5.64) (19.28)
Pandharpuri 10.21a 6.05a 52.31a 31.86a 0.61a
(12.02) (14.15) (11.71) (6.18) (22.21)
Tarai 9.69c 5.57b 50.26b 31.48b 0.58b
(9.82) (12.32) (7.37) (3.31) (20.82)
Nili-Ravi 9.33d 5.48c 47.38d 30.74d 0.60a
(9.83) (13.63) (8.88) (4.44) (20.79)
Surti 9.97b 5.55b 50.45b 31.17c 0.57b
(9.34) (12.7) (8.38) (4.2) (20.23)
Jaffrabadi 9.55c 5.65b 49.79b 31.14c 0.60a
(11.05) (13.26) (8.52) (4.4) (21.06)
CD at 5% 0.170 0.106 0.855 0.278 0.018
Each value is a mean of 500 sperm from five bulls of each breed.
*
Values within columns with different superscripts differ (P < 0.05).
**
Values in parenthesis represent coefficient of variation.

values, followed by Surti and Tarai, then Bhadawari and Table 2

Mehsana and finally Murrah. Biometry of the sperm tail in Indian buffaloes
Breed Length of Width of Length of
3.2. Sperm tail biometry midpiece (mm) midpiece (mm) tail (mm)
Bhadawari 16.12b* 1.03d 54.23b
Midpiece length (Table 2) was greatest in Surti, (9.45)** (11.08) (8.34)
Mehsana and Murrah, followed by Bhadawari; Tarai,
Mehsana 16.44a 1.08c 54.76b
Nili-Ravi and Jafrabadi were similar, whereas Pand-
(8.23) (9.78) (9.23)
harpuri had the least value (14.73 mm). The width of the
midpiece was 1.35 mm in Pandharpuri followed by Murrah 16.36a 1.11b 57.02a
(7.62) (11.35) (7.83)
Murrah, Mehsana and Bhadawari; values for each of
these four breeds differed significantly from each other, Pandharpuri 14.73d 1.35a 52.96c
(14.91) (16.21) (12.79)
whereas the midpiece width was least in Tarai, Nili-
Ravi, Jaffrabadi and Surti breeds. Based on tail length, Tarai 15.78c 1.00e 54.47b
the breeds could be classified into four groups; Murrah (8.32) (8.21) (7.97)
(57.02 mm) and Jaffrabadi (56.96 mm) were largest, Nili-Ravi 15.56c 0.94e 54.57b
followed by Bhadawari, Mehsana, Tarai, Nili-Ravi and (9.32) (9.09) (7.88)
Surti (54.23–55.17 mm), with the shortest tail length in Surti 16.57a 0.95e 55.17b
Pandharpuri (52.96 mm). (8.99) (11.75) (9.36)
Jaffrabadi 15.67c 0.99e 56.96a
3.3. Multivariate analysis and clustering (10.88) (11.02) (9.05)
CD at 5% 0.30 0.02 0.98
The multivariate analysis (PCA) for the eight sperm
Each value is a mean of 500 sperm from five bulls of each breed.
morphometry end points in the eight breeds indicated *
Values within columns with different superscripts differ
that the percent variations explained by the first three (P < 0.05).
**
principal components were 77.08, 17.50 and 3.29, Values in parenthesis represent coefficient of variation.
 R.A.K. Aggarwal et al. / Theriogenology 68 (2007) 682–686
Fig. 1. Scatter graph based on principal component analysis of eight
sperm biometry characters, showing clustering of eight buffalo breeds
(1, Bhadawari; 2, Mehsana; 3, Murrah; 4, Pandharpuri; 5, Tarai; 6,
Nili-Ravi; 7, Surti and 8, Jaffrabadi). [PCA I = 77.08%, PCA
II = 17.50% and PCA III = 3.29%].
respectively. The three-dimensional scatter graph
plotted the eight breeds in three clusters (Fig. 1).
Bhadawari, Mehsana, Tarai, Nili-Ravi, Surti, Jaffrabadi
breeds were grouped in one cluster but Murrah,
although a different identity, appeared near this cluster.
The Pandharpuri breed had a distinctly different
identity.
4. Discussion
We determined average values, along with their
corresponding CD, for eight sperm biometry end points
in eight breeds of Indian buffaloes (Bubalus bubalis).
There were several significant differences among breeds
for each end point. However, multivariate analysis
grouped these eight breeds into three clusters/identities;
Bhadawari, Mehsana, Nili-Ravi, Tarai, Surti and
Jaffrabadi clustered together (within this cluster, Tarai,
Surti and Jaffrabadi were highly alike). It was
noteworthy that these three breeds were in the same
group for many of the individual end points (univariate
analysis). In contrast, Murrah and Pandharpuri were
separate identities in the three-dimensional scatter
graph, consistent with the univariate analysis, where
they were in opposing groups for most end points.
There are some previous reports [18–21] regarding
sperm biometry of Murrah and Surti buffalo bulls.
Kodagali et al. [18], studied sperm biometry in
Surti breed of buffalo and reported measurements
of head length (7.618  0.22 mm), head width
(4.591  0.51 mm), head area (27.210  2.78 mm2),
midpiece length (0.837  0.09 mm) and tail length
(56.715  0.93 mm); which were not different in
subfertile bulls. However, sperm head width at base
(2.016  0.16 mm) was significantly lower in subfertile
bulls versus normal bulls (2.589  0.21 mm). Hara-
685
panhalli and Mukherjee [19], while studying the effect
of different diluents on the sperm cytomorphology of
Murrah buffalo, reported that preservation of sperm in
Egg-Yolk-Glucose-Sodium Citrate (EYGC) diluent for
72 h did not alter the head width in comparison to its
value in sperm of neat semen. In neat semen, head
length, head width, head area, midpiece length and
midpiece width were 7.323 mm, 4.903 mm,
28.960 mm2, 11.288 mm and 0.551 mm, respectively.
Venkataswami and Vedanayagam [20] reported the
head length and width of Murrah buffalo sperm as
7.40  1.07 mm and 4.48  1.29 mm, respectively,
whereas length of midpiece and tail were
12.41  0.85 and 43.61  5.56 mm, respectively. They
suggested that these values were higher in bulls that
were >10 years old. Tomar et al. [21] analyzed seven
semen samples of Murrah buffalo bulls in different
diluents and found that average length and width of
sperm head did not vary significantly; the average
length and width of the normal sperm was
7.656  0.060 mm and 5.003  0.84 mm, respectively.
Our results on different parameters of buffalo sperms
seemed slightly higher than reported in earlier studies as
described above; the previous studies were done for
different purposes and on small number of samples
using microscope with ocular/stage micrometer for
measurement. In contrast, we have used large number of
samples for measuring biometry using automated
instruments, so as to reduce error.
Sperm morphometry has been correlated with
fertility in stallion [11,22] and boars [23]. It has been
postulated that sperm head area and shape effects total
sperm volume and in turn sperm freezability by
influencing sperm cryoresistance [24]. Thus, sperm
biometry data of buffalo bulls belonging to diverse
breeds, generated in this study, may form the basis for
correlating the fertility status.
This study reports biometry of live sperm for eight
breeds of Indian buffaloes (Bubalus bubalis) and
multivariate analysis grouped them in three separate
clusters. These data provide a preliminary basis for
assessing sires with better fertilizing potential in
combination with other morphological and reproductive
traits.
Acknowledgements
The authors thank Livestock Research Station
Vallabhnagar (Gujarat); Semen production center
(Nabha) under Livestock Development Board (Pun-
jab); Gujarat Agricultural University, Junagarh
(Gujarat); Artificial Breeding Complex, NDRI, Karnal;
686 R.A.K. Aggarwal et al. / Theriogenology 68 (2007) 682–686
Dudhsagar Dairy, Mehsana (Gujarat) for supplying the
semen of Surti, Nili-Ravi, Jaffrabadi, Murrah and
Mehsana buffalo breeds, respectively, for use in the
present study. Thanks are also due to Sh. Gian Singh of
the computer center, NDRI, Karnal for help with
statistical analysis of data. We are also thankful to Dr.
R.K. Vijh, Principal Scientist; Dr. P.K. Singh and Dr.
B.P. Mishra, Senior Scientists of NBAGR, who helped
in interpretation of statistical data and gave valuable
inputs in preparation of this manuscript.
References
[1] Brahmkshtri BP, Edwin MJ, John MC, Nainar AM, Krishnan
AR. Relative efficacy of conventional sperm parameters and
sperm penetration bioassay to assess bull fertility in vitro. Anim
Reprod Sci 1999;54:159–68.
[2] Correa JR, Pace MM, Zavos PM. Relationships among frozen–
thawed sperm characteristics assessed via the routine semen
analysis, sperm functional tests and fertility of bulls in an
artificial insemination program. Theriogenology 1997;48:
721–31.
[3] Rodriguez-Martinez H, Larsson B. Assessment of sperm ferti-
lizing ability in farm animals. Acta Agric Scand Suppl 1998;29:
12–8.
[4] Zhang BR, Larsson B, Lundeheim N, Rodriguez-Martinez H.
Sperm characteristics and zona pellucida binding in relation to
fiedd fertility of frozen thawed semen for dairy AI bulls. Int J
Androl 1998;21:207–16.
[5] Budworth PR, Amann RP, Chapman PL. Relationships between
computerized measurements of frozen and thawed bull sperma-
tozoa and fertility. J Androl 1982;9:41–54.
[6] Farrell PB, Presicce GA, Brockett CC, Foote RH. Quantification
of bull sperm characteristics measured by computer-assisted
sperm analysis (CASA) and the relationship to fertility. Ther-
iogenology 1998;49:871–9.
[7] Januskauskas A, Gil J, Soderquist L, Rodriguez-Martinez H.
Assessment of sperm characteristics post-thaw and response to
calcium ionophore in relation to fertility in Swedish dairy AI
bulls. Theriogenology 2000;53:859–75.
[8] Kjaestad H, Ropstad E, Berg KA. Evaluation of spermatological
parameters used to predict the fertility of frozen bull semen. Acta
Vet Scand 1993;34:299–303.
[9] Killian GJ, Chapman DA, Rogowski LA. Fertility-associated
proteins in Holstein bull seminal plasma. Biol Reprod 1993;49:
1202–7.
[10] Larson JL, Miller DJ. Can relative spermatozoal galactosyltrans-
ferase activity be predictive of dairy bull fertility? J Dairy Sci
2000;83:2473–9.
[11] Casey PJ, Gravance CG, Davis RO, Chabot DD, Liu IKM.
Morphometric differences in sperm head dimensions of fertile
and subfertile stallions. Theriogenology 1997;47:575–82.
[12] Acharya RM, Bhat PN. Livestock and poultry genetic resources
of India. Res Bull 1984;1:11–3.
[13] Mahadevan P. Water buffalo research: possible future trends.
World Anim Rev 1978;25:2–7.
[14] Semen Bank. Laboratory techniques for evaluation of frozen
semen. Chennai: Madras Veterinary College, Tamil Nadu Veter-
inary and Animal Sciences University; 2002.
[15] Hancock JL. A staining technique for the study of temperature
shock in semen. Nature 1951;167:323–4.
[16] Harvey WR. Mixed model least square and maximum likelihood
computer programme PG-II Version, 1990.
[17] Panse VG, Sukhatme PV. Statistical methods for agricultural
workers, Revised ed., New Delhi: Indian Council of Agricultural
Research; 1985.
[18] Kodagali SB, Bhavsar BK, Deshpande AD. Biometrics of Surti
buffalo spermatozoa. Indian Vet J 1973;50:50–4.
[19] Harapanhalli MD, Mukherjee DP. Effect of different diluents on
the cytomorphology of buffalo spermatozoa. Indian J Anim Prod
1973;4:128–35.
[20] Venkataswami V, Vedanayagam AR. Biometrics of spermatozoa
of cattle and buffaloes. Indian Vet J 1962;39:287–91.
[21] Tomar NS, Pande R, Desai RN. Efficacy of the semen diluents to
preserve the normal morphology of bovine spermatozoa. Indian J
Dairy Sci 1964;17:104–6.
[22] Gravance CG, Liu IKM, Davis RO, Hughes JP, Casey PJ.
Quantification of normal head morphometry of stallion sperma-
tozoa. J Reprod Fertil 1996;108:41–6.
[23] Hirai M, Boersma A, Hoeflich A, Wolf E, Foil J, Aumuller R,
et al. Objectively measured sperm motility and sperm head
morphometry in boars (Sus scrofa): relation to fertility
and seminal plasma growth factors. J Androl 2001;22:104–10.
[24] Esteso MC, Soler AJ, Fernandez-Santos MR, Quintero-Moreno
AA, Garde JJ. Functional significance of the sperm head mor-
phometric size and shape for determining freezability in Iberian
Red Deer (Cervus elaphus hispanicus) epididymal sperm sam-
ples. J Androl 2006;27:662–70.