Original Paper
Received: February 16, 2000
Gynecol Obstet Invest 2002;53:6–12
Labor in the Grand Multipara
Sven Lyrenäs
Department of Women’s and Children’s Health, Obstetrics and Gynaecology, Uppsala University,
University Hospital, Uppsala, Sweden
Key Words
Labor W Active phase W Dystocia W Birth weight W Grand
multipara
Abstract
Objective: To assess delivery time and the frequency of
dystocia in grand multiparous (GMP) and grand grand
multiparous (GGMP) women. Method: Labor records
from 272 women with parity 4–8 (GMP) and 56 women
with parity 9 or higher (GGMP) were retrospectively
reviewed regarding duration of labor and the occurrence
of dystocia. As a comparison, data from 263 women with
parity 1–3 (LMP) and 87 nulliparas (NP) was used. Re-
sults: Duration of labor, defined as time from admission
of the laboring woman to delivery of the infant, lasted in
median 2.0, 2.3 and 3.1 h in LMP, GMP and GGMP,
respectively. There was no difference in mean cervical
dilatation on admission in multiparous women. In
GGMP, the presenting part was more frequently posi-
tioned above the pelvic inlet at time of admission. In mul-
tiparous women admitted during the latent phase, the
active phase of labor lasted in median 3.7 h in GMP and
4.7 h in GGMP, significantly longer than 2.9 h in LMP.
During the active phase of labor, GGMP experienced
arrest of cervical dilatation more frequently than the
LMP. In parous women, there was a positive relationship
between parity and duration of the active phase of labor
as well as the duration of labor. Infant’s birth weight
© 2002 S. Karger AG, Basel
ABC 0378–7346/02/0531–0006$18.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/journals/goi
Accepted after revision: July 12, 2001
increased by parity. Conclusion: Duration of the active
phase of labor increased after the fourth child. Failure of
descent of the presenting part during the first stage of
labor in addition to arrest of cervical dilatation was asso-
ciated with a high cesarean section rate in the GMP
woman.
Copyright © 2002 S. Karger AG, Basel
Introduction
The risk of complications during a first pregnancy and
labor exceeds that of the second and third [1]. For exam-
ple, dystocia, or abnormal labor, is a major problem found
mainly in the first labor [1]. Some authors claim, however,
that maternal and fetal risks are increased after the fourth
delivery [2–5]. Placental abnormalities [3], increased fre-
quencies of fetal malpresentation [3, 4] and macrosomia
[5] are reported reasons for this.
Dystocia is another possible explanation of increased
maternal and neonatal morbidity in relation to many pre-
vious deliveries. Duration of labor and frequency of labor
disturbances were compared in this study between wom-
en with very high, high and low parity. The study was car-
ried out in a hospital with a high incidence of young grand
multiparas. The short period over which the study was
performed should prevent possible bias due to change,
over time, in obstetric practice.
Sven Lyrenäs
Department of Women’s and Children’s Health, Obstetrics and Gynaecology
Akademiska Sjukhuset
SE–751 85 Uppsala (Sweden)
128.59.222.107 - 11/30/2017 1:01:07 AM
Tel. +46 18 6110000, Fax +46 18 157855, E-Mail Sven.Lyrenas@kk.uas.lul.se
Columbia University
Downloaded by:
 Material and Methods test, when appropriate. For correlation analysis, Pearson correlation
coefficient was used. Statistical significance was determined at the
Data was collected from medical records retrospectively. The p ! 0.05 level.
investigation included files from 678 women, scheduled for vaginal
delivery during a 4-month period (November 1987–February 1988)
in the King Khaled Hospital (KKH), Najran, Saudi Arabia. The
Results
KKH is a modern hospital and the delivery unit is equipped with
facilities such as electronic fetal monitoring and ultrasonography. An
efficient outpatient unit with an antenatal clinic had been opened the The main characteristics of the women in the four
preceding year. Ultrasound screening to ascertain gestational age was groups are given in table 2. Thirteen (4.0%) of the women
not done. who delivered for the fifth time or more had a previous
A parity number was given to each delivered woman, not count-
ing the current delivery. The women were divided into four groups
according to their parity number: (1) nulliparas (NP) with no pre-
vious delivery; (2) low multiparas (LMP) with 1–3 previous deliver-
ies; (3) grand multiparas (GMP) with 4–8 previous deliveries and Table 1. Parity distribution in the four groups, 678 deliveries
(4) grand grand multiparas (GGMP) with 9 or more previous deliver-
ies. The distribution of women according to parity number is shown
Group Parity n Frequency Cum. freq. Number per
in table 1.
% % group (%)
Duration of labor was defined according to O’Driscoll and
Maegher [6] as the time the laboring mother spent in the delivery unit
NP 0 87 12.8 12.8 87 (12.8%)
from the time of admission until delivery of her infant. Labor infor-
mation was collected from the woman’s partogram, the routine labor LMP 1 97 14.3 27.1
recording method used by doctors and midwives. The first stage was 2 87 12.8 40.0
subdivided into two phases, latent phase and active phase. The latent 3 79 11.7 51.8 263 (38.8%)
phase was defined as the time from admission until the first measure-
GMP 4 72 10.6 62.4
ment of 4 cm cervical dilatation. The active phase was estimated
5 75 11.1 73.5
from first measurement of 4 cm cervical dilatation to complete cervi-
6 59 8.7 82.2
cal dilatation. The second stage of labor lasted from first measure-
7 34 5.0 87.2
ment of complete cervical dilatation to delivery of the baby. The
8 32 4.7 91.7 272 (40.1%)
third stage of labor was not included in this study. Protracted cervical
dilatation was present when dilatation rate was ! 1.2 cm/h. Second- GGMP 9 21 3.1 94.8
ary arrest of cervical dilatation was present when no dilatation was 10 24 3.5 98.6
recorded during 2 h or more. Failure of descent during the first stage 11 8 1.2 99.6
was present when the presenting part remained at or above the pelvic 12 1 0.1 99.7
inlet until an intervention was performed. 13 1 0.1 99.9
16 1 0.1 100 56 (8.3%)
Statistical Analyses
Differences in duration of labor were analyzed with the Mann- NP = Nullipara, LMP = low multipara, GMP = grand multipara,
Whitney U-test. Student’s t-test was used for comparisons of means. GGMP = grand grand multipara.
Differences in frequencies were assessed by ¯2-test or Fischer’s exact

Table 2. Main characteristics and obstetric outcome

NP LMP GMP GGMP

(n = 87) (n = 263) (n = 272) (n = 56)

Age, years, mean (BSD) 20.9 (4.6) 24.4 (4.4) 29.8 (4.6) 34.5 (5.5)
Previous cesarean delivery, n (%) 16 (6.1) 10 (3.7) 3 (5.4)
Artificial rupture of membranes, n (%) 11 (12.6) 26 (9.8) 35 (12.9) 7 (12.5)
Labor enhancement with oxytocin, n (%) 36 (41.4) 42 (15.6) 49 (18.1) 4 (7.1)
Spontaneous vaginal delivery, n (%) 67 (77.0) 247 (93.9) 252 (92.6) 50 (89.3)
Vacuum extraction or forceps, n (%) 14 (16.1) 5 (1.9) 6 (2.2) 1 (1.8)
VBAC frequency, n (%) 13 (81.3) 7 (70.0) 3 (100)
Cesarean section, n (%) 6 (6.9) 11 (4.2) 14 (5.2) 5 (8.9)

VBAC = Vaginal birth after cesarean section.

Labor in the Grand Multipara Gynecol Obstet Invest 2002;53:6–12 7

128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by:
Fig. 1. Cervical dilatation on admission.
Frequency (%) of women admitted during
the latent phase of first stage of labor (right
panel) and during the second stage of labor
(left panel).
Fig. 2. Position of the presenting part (PP)
on admission. Frequency (%) of women ad-
mitted with PP at or above pelvic inlet (left
panel) and with PP at or below spinal level
(right panel). * p ! 0.05, ** p ! 0.01 vs. LMP
group as determined by ¯2-test.
cesarean delivery. Ten of these women (76.9%) had a suc-
cessful vaginal birth after cesarean section (VBAC). The
total VBAC frequency after one previous cesarean section
was 22 out of 23 (95.7%). The cesarean section rates were
not significantly different in the three multiparous groups
(table 2). The overall cesarean section rate was 5.3%.
There were six repeat operations, while in 30 women the
cesarean delivery was the woman’s first operative de-
livery.
The phase of labor and the position of the presenting
part, on admission, are shown in figures 1 and 2. In the
multiparous groups, an equal number were admitted dur-
ing the latent phase of labor, namely 41, 37 and 39% for
8 Gynecol Obstet Invest 2002;53:6–12
LMP, GMP and GGMP, respectively. On admission,
there was no difference in mean cervical dilatation be-
tween LMP, GMP and GGMP. Eighteen percent of the
GMP and GGMP mothers were admitted during the sec-
ond stage of labor, not significantly different from 20% in
the LMP group (fig. 1). In the GGMP, the presenting part
was more frequently positioned at or above the pelvic
inlet on admission than in GMP and LMP (p ! 0.05 and
p ! 0.01, respectively) (fig. 2). Furthermore, on admis-
sion, the presenting part was seldom positioned at or
below the spinal level (fig. 2) in both GMP and GGMP
compared to LMP (p ! 0.01 and p ! 0.05, respectively)
(fig. 2).
Lyrenäs
128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by:
Fig. 3. Cumulative distributions of the dura-
tion of active phase of labor in 38 nullipara
(NP), 105 para 1–3 (LMP), 105 para 4–8
(GMP) and in 18 para 9–16 (GGMP). The
active phase was estimated from the first
measurement of 4 cm cervical dilatation to
complete cervical dilatation.

Table 3. Duration of labor (median)

NP LMP GMP GGMP

Normal first stage of labor, n (%) 44 (53.7)*** 204 (80.0) 197 (76.1) 32 (62.7)*
Active phase, h 5.0 (1.0–22.0)* 2.9 (0.1–24.0) 3.7 (0.1–25.5)* 4.7 (0.6–36.0)*
(min-max) n = 38 n = 105 n = 105 n = 18
Second stage of labor, h 0.5 (0.1–7.3)*** 0.2 (0.1–4.5) 0.2 (0.1–3.5) 0.2 (0.1–4.0)
(min-max) n = 56 n = 179 n = 185 n = 36
Duration of labor, h 5.8 (0.1–30.0)*** 2.0 (0.1–28.8) 2.3 (0.1–36.0) 3.1 (0.1–36.1)
(min-max) n = 87 n = 263 n = 272 n = 56

* p ! 0.05, ** p ! 0.01, *** p ! 0.001 vs. LMP group as determined by ¯2-test or by Mann-Whitney U-test.

Active Phase of Labor more often exceeded 12 h in GGMP (p = 0.01) and NP

In each of the three multiparous groups, the active
phase of labor was possible to assess in approximately
40% of the women. These subgroups of women were all
admitted during the latent phase of labor (fig. 1, table 3).
The median duration of the active phase of labor was lon-
ger in GMP (p = 0.035) and in GGMP (p = 0.031) than in
LMP. In nulliparous women, the active phase was longer
than in LMP (p = 0.01) but not significantly longer than in
GMP and GGMP. The cumulative frequency of the
active phase of labor is shown in figure 3. The duration of
the active phase more often exceeded 5 h in GGMP wom-
en than in LMP (p ! 0.05). Furthermore, the active phase
(p ! 0.05) than in LMP (fig. 3). The GGMP women had
their active phase of labor more often disturbed by an
arrest of cervical dilatation as compared to LMP (p !
0.01) and GMP (p ! 0.01). Failure of descent during first
stage occurred more frequently in the GMP than in the
LMP (p = 0.03). This difference was not statistically sig-
nificant in the GGMP (p = 0.07). In 11 multiparous wom-
en with failure of descent during first stage of labor in
addition to secondary arrest of cervical dilatation, the
cesarean section rate was 55% (table 4). In this group, the
median infant birth weight in 6 women having a cesarean
delivery (3,275 g; min-max 2,700–3,650) did not differ in

Labor in the Grand Multipara Gynecol Obstet Invest 2002;53:6–12 9

128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by:
Fig. 4. Cumulative distributions of the dura-
tion of labor in 87 nullipara (NP), 263 para
1–3 (LMP), 272 para 4–8 (GMP) and in 56
para 9–16 (GGMP). Duration of labor was
defined as the time the laboring mother
spent in the delivery ward from time of
admission until delivery of her infant.

Table 4. Obstetric outcome in 84 women with failure of descent of the presenting part during normal and dysfunc-
tional first stage labor

Normal first stage labor Arrest of cervical Protracted cervical

(n = 40) dilatation (n = 11) dilatation (n = 33)
vaginal cesarean vaginal cesarean vaginal cesarean

NP (n = 9) 3 5 1
LMP (n = 25) 13 1 2 1 5 3
GMP (n = 41) 17 3 3 2 14 2
GGMP (n = 9) 3 3 3
Total, % 36 (90.0) 4 (10.0) 5 (45.5) 6 (54.5)**,+ 27 (81.8) 6 (18.2)

** p ! 0.01 vs. normal first stage labor as determined by Fisher’s exact test.
+ p ! 0.05 vs. protracted cervical dilatation as determined by Fisher’s exact test.

comparison to that of the 5 women who delivered vaginal- Duration of Labor

ly (3,200 g; min-max 2,600–3,800).
Second Stage of Labor
The median second stage of labor lasted 0.2 h in all
three multiparous groups. The NP had longer second
stage labors than LMP, GMP and GGMP (p ! 0.001, p !
0.001 and p = 0.001, respectively). Between the multipa-
rous groups there were no significant differences regard-
ing duration of second stage or the frequency of pro-
tracted second stage. The frequency of forceps and vacu-
um deliveries was equally low in the multiparous groups,
approximately 2%.
The median duration of labor, assessed from the time
of admission of the woman until delivery of her infant,
lasted 2.0 h for LMP, 2.3 h for GMP and 3.1 h for GGMP.
The difference between GGMP and LMP was not statisti-
cally significant (p = 0.16). Figure 4 shows the cumulative
frequencies of duration of labor. Fewer GGMP had deliv-
ered after 4, 8 and 12 h than had LMP (p ! 0.02, p ! 0.05
and p ! 0.05, respectively). For GMP, the same difference
in comparison with LMP reached statistical significance
only at 12 h (p ! 0.05). Duration of labor exceeded 16 h in
8.0, 2.3, 4.4 and 5.2% of the NP, LMP, GMP and GGMP,
respectively (fig. 4).

10 Gynecol Obstet Invest 2002;53:6–12 Lyrenäs

128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by:
Table 5. Fetal weight and Apgar score at
5 min (mean B SD)
Fetal weight, g
Apgar score 5 min
** p ! 0.01 vs. LMP group as determined by t-test.
+ p ! 0.05 vs. GMP group as determined by t-test.
In multiparous women, duration of the active phase of
labor and duration of labor increased by parity (rp =
0.255, p ! 0.001 and rp = 0.122, p = 0.003, respectively).
Oxytocin for labor enhancement was given to 16% of the
LMP, 18% of the GMP and to 7% of the GGMP (table 2).
The differences were not significant. The mean birth
weights of the infants are shown in table 5. The differ-
ences in mean birth weight between GGMP and LMP and
GGMP and GMP were statistically significant (p ! 0.01
and p ! 0.05). Infants’ birth weight increased by parity
(rp = 0.221, p ! 0.001). The woman who gave birth to her
17th baby delivered the heaviest infant in the study
(4,750 g). The stillbirth ratio was close to 1% in all four
groups and 1.0% in all the 678 deliveries.
Discussion
During the last decades, an increased interest in the
GMP woman is mirrored in a number of articles address-
ing this topic ‘the dangerous multipara’ [4, 5, 7–10]. Most
of these studies have confirmed an increased risk of ane-
mia, hypertension, preeclampsia, placenta praevia, uter-
ine rupture, abnormal presentation and operative deliv-
ery in pregnant women with high parity [4, 8, 10]. High
parity has consequently been related to an increased neo-
natal morbidity and perinatal mortality, often combined
with low socioeconomic status and increasing age [11, 12].
The present investigation was carried out in a rapidly
developing country with no major economic problems
where the government financially supported large fami-
lies. Due to religious reasons, female smoking was pro-
hibited and uncommon. Alcoholic beverages in any form
were not permitted. In Sweden, Lyrenäs et al. [13] found
the mean age of the Swedish primipara to be 25.4 years.
The mean age of the GMP (para 4–8) in the present study
was 29.8, only 4 years older. Socioeconomic factors and
age can thus barely contribute to a major bias in the
present study. The delivery unit equipment was compara-
ble to that found in western developed countries. In the
Labor in the Grand Multipara
NP LMP GMP GGMP
2,908B358** 3,078B525 3,200B508 3,350B528**,+
9.3B1.6 9.6B1.5 9.6B1.4 9.5B1.7
KKH, the women labored without any form of conduc-
tion analgesia, no pain relief other than nitrous oxide and,
occasionally, opiates were offered. The husband or other
relatives were not allowed to visit the labor room, conse-
quently, the mother’s emotional support was given by the
midwife and the obstetrician.
The obstetric course of the GMP regarding duration
and disturbances of labor has been scantily reported in
previous studies. On the contrary, there are many reports
on the NPs’ labor [1]. During the first delivery, there is an
increased risk of dystocia and operative delivery as com-
pared to the following three or four labors [1]. The present
study indicates that after nine deliveries a woman re-
sumes a similar risk of dystocia as during her first deliv-
ery. Sharfman and Silverstein [14] reviewed the files of
403 women with parity 7 or higher. In their group, 45%
were delivered within 6 h while in 23% labor exceeded
10 h. The corresponding finding in the present study was
74 and 14%, respectively. Evaldson [7] reported a total
frequency of disturbances during labor of 36.7% in primi-
paras, 22.2% in para 2–5 and 35.7% in para 6. The similar
frequencies for the NP and the GMP regarding dystocia
are in agreement with the present study.
The GMP women were not admitted earlier or later
during the first stage than the LMP. According to Hen-
dricks et al. [15] and to Juntunen and Kirkinen [16], a
difference in duration of labor between NPs and GMPs
lies mainly in the latent phase [15, 16] and the second
stage [15] of labor. The present study indicates a differ-
ence in the active phase since this phase was longer in
GMP and GGMP than in LMP but not significantly dif-
ferent from the NP. The mechanism behind this finding is
not clear. Hypothetically, a reduced contractive force of
the myometrium in the GMP may follow after numerous
previous labors. In GMPs with protracted first stage
labor, the woman usually resumed normal progress after
receiving intravenous fluid of glucose and, occasionally,
amniotomy. Secondary arrest of cervical dilatation, which
has been reported as a major cause of dystocia in nullipa-
rous women [1], was also more common in the GGMP
Gynecol Obstet Invest 2002;53:6–12 11
128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by:
group. In GMP and GGMP, descent of the presenting
part was often absent during the first stage of labor. When
amniotomy was performed after full cervical dilatation
and the preceding first stage of labor was normal or pro-
tracted, a rapid progress with descent of the presenting
part and vaginal delivery followed in most cases (table 4).
In contrast to Juntunen and Kirkinen [16], who reported a
similar observation, the frequency of deep occiput trans-
verse positions in GMP and GGMP were low. If, instead,
labor had been accompanied by an arrest of cervical dila-
tation in addition to failure of descent, only 36% of the
women were delivered vaginally after amniotomy (ta-
ble 4). In the complete material, 19 out of 36 cesarean sec-
tions (53%) were carried out in laboring women with fail-
ure of descent.
Cesarean delivery was rarely the woman’s choice in the
study hospital. Trial of labor was instead the mode of
practice after two and even more previous operative
deliveries. Consequently, all 3 GGMP who were allowed
trial of labor after a previous single cesarean birth deliv-
ered vaginally.
The present study indicates that the GMP woman will
face an increased frequency of dystocia during the first
stage of labor such as arrest of cervical dilatation and fail-
ure of descent of presenting part. Factors that have been
associated with the progress of labor are epidural analge-
sia [17], oxytocin [18], fetal weight [19] and the presence
of emotional support during labor [6]. The absence of con-
duction analgesia and an equal frequency of oxytocin use
in labor enhancement will minimize the bias in this study.
It cannot be ruled out that a possible fear of labor
enhancement in the GMP and GGMP may have contrib-
uted to the equal use of oxytocin. Consequently, 1 of the 9
GGMP with failure of descent had labor enhanced by
oxytocin. Furthermore, only 1 of the 11 GGMP with pro-
tracted first stage labor received oxytocin. This circum-
stance together with increasing birth weights in GMP and
in GGMP may, to some extent, have contributed to the
longer active phase of labor found in these two groups.
In conclusion, the present study indicates a successive
prolonging of labor by parity in multiparous women. A
combination of factors, some of which are yet unknown,
causing inefficient uterine contractions together with in-
creasing birth weight of infants born to high parity women
are the most probable contributors to the recorded distur-
bances of GMPs’ labor. Failure of descent of the present-
ing part during the first stage of labor in addition to a sec-
ondary arrest of cervical dilatation was associated with a
high cesarean section rate.
Acknowledgments
This study could not have been carried out without the assistance
of Dr. B. Svanberg, Dr. P. Kasturi, Dr. A. Watfa and Miss C. Creed,
which is acknowledged and appreciated.

References

1 Boylan PC: Labor in the primigravid patient.
Curr Probl Obstet Gynecol Fertil 1991;14:5–
34.
2 Solomon B: The dangerous multipara. Lancet
1934;ii:8–11.
3 Israel SL, Blazar AS: Obstetric behavior of the
grand multipara. Am J Obstet Gynecol 1965;
91:326–332.
4 Chang A, Larkin P, Esler EJ, Condie R, Morri-
son J: The obstetric performance of the grand
multipara. Med J Aust 1977;i:330–332.
5 Abu-Heija AT, Chalabi HE: Great grand multi-
parity: Is it a risk? Int J Gynecol Obstet 1997;
59:213–216.
6 O'Driscoll K, Maegher D: Active Management
of Labor. Philadelphia, Saunders, 1980, pp 19–
22.
7 Evaldson G: The grand multipara in modern
obstetrics. Gynecol Obstet Invest 1990;30:
217–223.
8 Tanbo TG, Bungum L: The grand multipara –
Maternal and neonatal complications. Acta
Obstet Gynecol Scand 1987;66:53–56.
9 Kjer JJ: Grand multipara. Zentbl Gynäkol
1989;III:1503–1507.
10 Brunner J, Melander E, Krook-Brandt M, Tho-
massen PA: Grand multiparity as an obstetric
risk factor: A prospective case-control study.
Eur J Obstet Gynecol Reprod Biol 1992;47:
201–205.
11 Mor-Yosef S, Seidman DS, Samueloff A,
Schenker JG: The effects of the socioeconomic
status on the perinatal outcome of grand multi-
para. Eur J Obstet Gynecol Reprod Biol 1990;
36:117–123.
12 Kilpatrick SJ, Laros RK: Characteristics of
normal labor. Obstet Gynecol 1989;74:85–87.
13 Lyrenäs S, Lutsch H, Hetta J, Lindberg B: Acu-
puncture before delivery: Effect on labor. Gy-
necol Obstet Invest 1987;24:217–224.
14 Scharfman E, Silverstein LM: The grand multi-
para. A survey of 403 cases. Am J Obstet Gyne-
col 1962;84:1442–1446.
15 Hendricks CH, Brenner WE, Kraus G: Normal
cervical dilatation pattern in late pregnancy
and labor. Am J Obstet Gynecol 1970;106:
1065–1082.
16 Juntunen K, Kirkinen P: Partogram of a grand
multipara: Different descent slope compared
with an ordinary parturitient. J Perinat Med
1994;22:213–218.
17 Thorp JA, Parisi VM, Boylan PC, Johnston
DA: The effect of continuous epidural analge-
sia on cesarean section for dystocia in nullipa-
rous women. Am J Obstet Gynecol 1989;161:
670–675.
18 Steer PJ, Carter MC, Beard RW: The effect of
oxytocin infusion on uterine activity levels in
slow labor. Br J Obstet Gynaecol 1985;92:
1120–1126.
19 Poma PA: Correlation of birth weights with
cesarean rates. Int J Gynecol Obstet 1999;65:
117–123.

12 Gynecol Obstet Invest 2002;53:6–12 Lyrenäs

128.59.222.107 - 11/30/2017 1:01:07 AM
Columbia University
Downloaded by: