Human Reproduction Vol.21, No.1 pp. 30–35, 2006
doi:10.1093/humrep/dei280
Advance Access publication September 9, 2005.
Number of germ cells and somatic cells in human fetal
ovaries during the first weeks after sex differentiation
E.Bendsen1,3, A.G.Byskov2, C.Yding Andersen2 and L.G.Westergaard1
1
Fertility Clinic, Department of Obstetrics and Gynaecology, University Hospital of Odense, Sdr. Boulevard 29, 5000 Odense C and
Laboratory of Reproductive Biology, Juliane Marie Center for Children, Women and Reproduction, University Hospital of Copenhagen,
Rigshospitalet, Blegdamsvej 9, 2100 København Ø, Denmark
2
3
To whom correspondence should be addressed. E-mail: [email protected]
BACKGROUND: This study presents the number of germ cells and somatic cells in human fetal ovaries during week
6 to week 9 post conception, i.e. the first weeks following sex differentiation of the gonads. METHODS: One ovary
with attached mesonephros from each of 11 individual legal abortions was used for estimation of cell numbers. After
recovery of the fetus, the ovary–mesonephric complexes were immediately isolated, fixed and processed for histology.
A stereological method was utilized to estimate the total number of oogonia in all ovaries and somatic cells in seven of
them. RESULTS: The number of oogonia per ovary increased from ~26 000 in week 6 to ~250 000 in week 9 and
somatic cells from ~240 000 to ~1.4ⴛ106. The ratio of oogonia to somatic cells tended to increase throughout the
period. The concentration of oogonia was similar in the cranial (mesonephric connected) part and the caudal part of
the ovaries. CONCLUSIONS: This is the first stereological estimation of the number of oogonia and somatic cells in
human fetal ovaries, and the first estimation of germ cells and somatic cells in ovaries aged <9 weeks. The number of
oogonia in week 9 is comparable to the numbers previously published based on non-stereological estimations. We
found early stages of meiosis in fetal ovaries from week 9.
Key words: fetal ovaries/first trimester/germ cells/human/in vivo
Introduction
The primordial germ cells originate in the yolk sac and migrate
to the gonadal anlage early in fetal life (Witschi, 1948; McKay
et al., 1953). The gonads form in close association with the
mesonephros that provide somatic cells to the ovary and the
testis (Byskov and Hoyer, 1994). Sexual differentiation begins
at ∼6 weeks post conception (p.c.) (Witschi, 1963; Wartenberg,
1982; Lovell-Badge and Robertson, 1990), governed by the
sex-determing region on the Y chromosome (Koopman et al.,
1990; Lovell-Badge and Robertson, 1990). In the female, the
absence of a Y chromosome will result in development of an
ovary. The fetal ovary is not capable of producing testosterone
or anti-Müllerian hormone (AMH) during this period, resulting
in degeneration of the Wolffian duct (lack of testosterone) and
growth of Müllerian duct derivatives (lack of AMH) (Winter
et al., 1977; Josso et al., 1993). The basis for further ovarian
development is founded early in embryonic life when proliferating oogonia and somatic cells populate the genital ridges and
where somatic cell–germ cell interaction is established. Since
all oogonia in the human ovary enter meiosis during fetal life,
putting an end to formation of additional oocytes, a reduction
in the number of oogonia would have deleterious effects on the
number of oocytes the girl is born with (Zuckermann, 1962).
The onset of meiosis is thus important in relation to the final
number of oocytes, but the time of this event has been variously
30
indicated as week 8 to week 13 (Gondos et al., 1986). This discrepancy is sometimes caused by use of menstrual age rather
than fetal age. Only one study reports on the number of germ
cells and somatic cells in fetal ovaries (Baker, 1963), and these
data are based on a total of only 14 ovaries, two being from
week 9 and the remaining 12 ovaries covering the two last trimesters. The interaction between germ cells and somatic cells
in the ovaries is crucial for survival and proliferation of the
germ cells, e.g. oocytes cannot develop without being enclosed
in a follicle (Hirshfield, 1991). Therefore, it is conceivable that
a certain numerical relationship between somatic cells, in particular granulosa cell precursors, and germ cells is needed for
normal ovarian development. So far, the number of somatic
cells has only been roughly estimated (Baker, 1963). It has
been proposed that the mesonephric cells, which begin the
gonadal invasion at the cranial end, have a positive effect on
the onset of meiosis, since meiosis is initiated in this area in
many mammalian species (Byskov, 1975).
The aim of the present study was to obtain precise estimates of
the number of oogonia and somatic cells and localize the first meiotic stages of the human first trimester ovaries. We have evaluated
11 + 30 ovaries (from 41 fetuses), isolated from legal abortions
performed during the last part of the first trimester of pregnancy.
The number of oogonia was estimated in 11 ovaries (from
fetuses aborted in weeks 6–9 inclusive) and somatic cells in seven
© The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
Germ and somatic cells in human fetal ovaries
of them by use of a stereological method, the optical fractionator
technique, which is known to provide precise and unbiased estimates of cell numbers in different organs (Gundersen et al., 1988).
The study of meiosis was performed on 30 ovaries (from
fetuses aborted in weeks 6–10 inclusive).
Materials and methods
Human fetal gonads
Human fetuses were obtained from women referred to the Department
of Gynaecology and Obstetrics, University Hospital of Odense,
Denmark, for legal induced abortion in the first trimester of pregnancy.The women wanted to end their pregnancy for social reasons.
They were healthy and had no history of medical treatment. In all
cases the women were informed orally and in writing about the aim
and procedures of the project, before they gave their consent to participate by signature. The Medical Ethics Committee in the counties of
Vejle and Fyn, Denmark, approved the project.
Operation technique
The operations were done according to the routine procedures in the
department, with slight modifications in order to prevent damage to the
fetus in utero, and carried no additional risk for the participating women
(Nauert and Freeman, 1994). In all abortions, the fetuses were monitored
with ultrasound equipment (Sonoline Prima; Siemens, Denmark). The
cervical canal was dilated, and a curette (Synevac Vacuum curette;
Berkeley Medevices Inc., USA) was inserted. A syringe (Plastipak;
Becton–Dickinson, USA) was connected to the curette and a vacuum
was applied manually. After recovery, the fetus was placed in a sterile
cup with culture medium (Dulbecco’s modified Eagles’s medium/Ham’s
F-12; GibcoBRL Life Technologies, Denmark). All handling of the tissue was performed under sterile conditions. The fetuses were dissected
under a stereomicroscope a few minutes after aspiration from the uterus.
The gonads were still present and isolated in ∼70% of the fetuses.
describing the nucleus of the cell). All sections were stained with haematoxylin and periodic acid–Schiff reagent.
Seven of the 11 ovaries used for counting cells were sectioned in the
longitudinal direction and four of the ovaries were sectioned transversely to the longitudinal direction (Figure 1A), which made it possible
to determine the topographic distribution of the oogonia in the ovaries.
The topographic distribution was illustrated by the variation in the concentration of oogonia between each of the serial sections, from the cranial, mesonephric-connected end to the caudal end of the ovaries.
Identification of cell types
The cells were defined by the morphology of the nuclei (Bendsen
et al., 2001).
Counting of cells
The number of oogonia was determined in all 11 ovaries. The number
of somatic cells was estimated in seven of the ovaries.
The total number of somatic cells counted encompasses all cells
that are not oogonia or blood cells.
Study of meiosis
All sections were examined for cells in meiosis.
Stereology
The number of cells was estimated using the optical fractionator technique (Gundersen et al., 1988), which provides precise and unbiased
Determination of fetal age
In this study, we only use true fetal ages, i.e. the time post conception
(p.c.), in contrast to gestational age, i.e. the time after the first day in
the last period. The fetal age was determined by measuring the length
of limbs and foot (Evtouchenko et al., 1996).
Determination of fetal sex
The chromosomal sex was determined by PCR technique (Nakahori
et al., 1991) and confirmed by the morphological appearance of the
histological sections.
Histology
One ovary–mesonephric complex from each of 11 fetuses was
included in the study of cell numbers and one ovary–mesonephric
complex from each of 30 fetuses was included in the study of meiosis.
The decision whether to use the right ovary or the left ovary was done
at random. However, the original position of one ovary could not be
determined since the organs in the fetal abdomen had been displaced
during the abortion.
The ovary–mesonephric complex was removed from the fetus in
toto and placed in Bouin’s fixative (Bie & Berntsen a/s, Denmark) for
up to 4 h, depending on the size of the complex and processed for paraffin embedding. The ovary–mesonephric complexes used for counting cells were cut into 30 μm thick serial sections (see Stereology) and
the ovary–mesonephric complexes used for the study of meiosis were
cut into 5 μm thick serial sections (this thickness is well-suited for
Figure 1. Human fetal ovaries.
31
E.Bendsen et al.
estimates of the total number of cells in an organ, by counting cells in
only a fraction of that organ (West and Gundersen, 1990; West et al.,
1991, 1996; Feinstein et al., 1996). The cells are counted in so-called
‘optical disectors’, based on parallel thin optical sections inside a
thick section (25–50 μm) of the ovary (see below). Even though the
concentration of cells in ovarian tissue is relatively large it is possible
to ‘look through’ these thick sections. The precision of this technique
is affected by variation in the thickness of the sections, i.e. CV (N)
(see below) will increase if the variation in the thickness is too large.
Our experience is that a section thickness of 30 μm is optimal for
human fetal ovarian tissue, i.e. a stable thickness from section to section is obtained. However, in the beginning of the study the ovaries
were cut into 50 μm thick sections and we found that the thickness
varied too much between sections in many of the ovaries. Therefore,
only three of these ovaries could be used in this study.
The Computer-Assisted-Stereological-Toolbox (CAST) grid system (Olympus A/S, Denmark) was used for all counting.
Sections
Cells were estimated from 10 to 15 sections that were selected at
equally spaced intervals along the entire extent of the ovary. The first
section was randomly selected between the first two to six sections,
depending of the size of the ovary, and thereafter every second to
sixth section was sampled. The fraction of sections sampled (ssf) was
consequently 1/2 to 1/6.
Sectional area: By positioning an unbiased counting frame of
known area, 134–668 μm2, at the coordinates of a rectangular lattice
superimposed on the section, a systematic random sample of the area
of each of the sections was achieved.
The fraction of the sectional area sampled (asf) = a (frame)/a (step).
Section thickness
The thickness of each of the sections used in the analysis was measured at every fifth point selected from the coordinates used to position
the disector samples, where cells were counted in the counting frame.
The thickness of these individual counting frames was weighed
according to the number of cells counted in the frame (tq) = 3(qi × ti)/3qi.
A known fraction of the thickness of the sections was sampled with
optical disectors at each position of the lattice. The counting frame
was moved a known distance (h), 10 μm, through the thickness of the
section.
The fraction of the section thickness sampled (tsf) = tq/h.
Counting
Optical disector counting rules (Gundersen et al., 1988; West and
Gundersen, 1990), based on the original physical disector counting rules
(Sterio, 1984; Gundersen, 1986), were used to count the number of cells
in the optical disectors. The counting unit was the nucleus of the cell.
Estimates
Estimates of the total number of cells in the gonad were calculated as
the product of the number of cells counted with the optical disectors
(3Q) and the reciprocals of the fraction of sections sampled (ssf), the
fraction of the sectional area sampled (asf), and the fraction of the
section thickness sampled (tsf).
N = 3Q×1/ssf×1/asf×1/tsf.
The efficiency of the fractionator is expressed by a coefficient of variation (CV) of the individual number estimates (N). CV depends on:
(i) homogeneity of the cell density in the ovary; (ii) variation in the
section thickness; and (iii) the number of sections sampled. On the
32
basis of similar analysis performed on other structures (Gundersen
and Jensen, 1987; West, 1993; West et al., 1996), we decided that a
CV (N) 0.1 would be adequate.
Counting of oogonia
The actual number of oogonia counted per ovary was 148–393 dispersed in 10–15 sections. CV (N) was 0.06–0.10.
Counting of somatic cells
The actual number of somatic cells counted per ovary was 186–433
dispersed in 10–15 sections. CV (N) was 0.06–0.10.
Results
Morphology of the ovaries
The length of the ovaries was ∼2 mm in week 6, and ∼3 mm in
week 10 (Figure 1C), but the ovaries became more rounded
during first trimester which may influence the actual measured
length. The diameter of the gonads increased from ∼200 μm in
week 6 to ∼400 μm in week 8. The homogeneous ovarian tissue
was intimately connected to the prominent mesonephros along
the whole length of the ovary. However, only at the cranial
end, dense cell streams—the so-called rete ovarii (Figure 1A
and C)—connected the cranially placed glomeruli with the
ovary, whereas no such connections were present caudally. In
the ovarian tissue, oogonia were scattered in all areas between
the somatic cells and a distinction between a germ cell-rich cortex and a medulla dominated by somatic cells was therefore not
obvious (Figure 1A and C). Oogonia in mitosis and oocytes in
early stages of meiosis tended to be organized in small groups
of 2–8 cells, which often displayed similar stages of cell cycle.
The nuclei of the oogonia were spherical with a diameter up to
10 mm, with one large, often spherical, nucleolus and smaller
ones (Figure 1B, D and E). The somatic cell nuclei were smaller
compared to the nuclei of oogonia and often elongated (Figure
1B, D and E). At all age stages, some oogonia were in mitosis
(Figure 1B and E). In the week 9 ovary, used for cell counting,
some germ cells had entered early stages of meiosis and were in
preleptotene and leptotene stage and were thus defined as
oocytes. This finding was confirmed by evaluating additionally
30 human ovaries cut into 5 μm thick sections (Figure 1D).
Meiotic stages were only observed in the cranial end of the
ovary, often close to the area that was connected to the mesonephros by the rete ovarii. These oocytes were included in the
counted numbers of oogonia, because the number of oocytes
constituted a fraction too small to justify separate counting.
Number of oogonia
In week 6 the number of oogonia per ovary ranged from ∼18 000
to ∼43 000 and increased to ∼250 000 in week 9 (Table I).
In the week 9 ovary, 2175 of the 253 600 oogonia were in
metaphase. In ovaries aged <9 weeks, fewer oogonia were in
metaphase. Only oogonia in metaphases were counted as mitosis.
Distribution of oogonia
The mean concentration, i.e. mean number of cells per counting frame, of oogonia was the same in the caudal and cranial
ends of the four ovaries that were cross-sectioned (Table II).
Germ and somatic cells in human fetal ovaries
Table I. Number of oogonia and somatic cells in human fetal ovaries
Fetal age p.c.
Fetal identification No. of No. of oogonia No. of
(weeks + days) number
oogonia in metaphase
somatic cells
6+5
6+5
6+6
6+6
6+6
7+3
7+4
8+1
8+2
8+5
9+5
1
2
3
4
5
6
7
8
9
10
11
18 451
21 181
42 934
20 435
26 737
39 228
37 180
82 230
67 137
114 944
253 600 2175
165 025
337 213
207 672
347 130
300 983
824 835
1 468 143
Thickness of section: most ovaries were cut in 30 μm thick sections, three in
50 μm thick sections (fetus numbers 3, 10 and 11).
The total number of somatic cells was not counted in fetus numbers 2, 5, 6
and 9.
p.c. = post conception.
Table II. Mean concentration of oogonia in the cranial end and in the caudal
end in human fetal ovaries
Fetal age
Fetal identification Mean concentration
(weeks + days) number
Cranial end
Caudal end
6+5
6+6
7+3
8+2
2
5
6
9
0.94 (sections 1–5)
0.71 (sections 1–6)
1.57 (sections 2–7)
1.35 (sections 1–6)
0.92 (sections 6–10)
0.74 (sections 7–13)
1.25 (sections 8–13)
1.13 (sections 7–12)
Mean concentration is the mean number of cells per counting frame.
Table III. Ratio of the number of oogonia (O) to the total number of somatic
cells (S)
Fetal age p.c. (weeks + days)
Fetal identification number
Ratio (O:S)
6+5
6+6
6+6
7+4
8+1
8+5
9+5
1
3
4
7
8
10
11
1:9
1:8
1:10
1:9
1:4
1:7
1:6
Number of somatic cells
In week 6 the number of somatic cells per ovary ranged from
∼165 000 to ∼337 000 and increased to ∼1 500 000 in week 9
(Table I).
The ratio between the number of oogonia and somatic cells
had a tendency to increase with age from around 1:8 in week 6
to 1:6 in week 9 (Table III).
Discussion
This is the first stereological estimation of the number of oogonia and somatic cells in sex-differentiated ovaries from first trimester fetuses. By analysing one ovary from each of 11 fetuses,
spanning a period in development from week 6 to week 9 p.c.,
we have shown that the number of oogonia gradually increases
from ∼26 000 to ∼250 000 at the end of this period. This information provides new data on the normal physiological development of the human ovary, during a sensitive period when sex
differentiation of the gonads occurs and when the function of
the organ for the rest of life is established. Therefore, the
present data represent the basis for further studies on the normal
function of the ovary, but also during conditions that may exert
negative effects on the ovary, such as drugs, cigarette smoke
and other compounds that interfere with growth and the delicate
hormonal balance during gonadal development.
Recently we have shown that 4-tert-octylphenol, one of the
environmental chemicals suspected to influence reproduction,
halves the number of male germ cells in human fetal testes cultured for 3 weeks in a concentration of 10 μmol/l, whereas no
effect could be seen on the ovaries (Bendsen et al., 2001).
Since no effect was detected in the ovaries it is possible that
proliferation of oogonia is not dependent on the same steroids.
To our knowledge, only two other studies have evaluated the
number of germ cells and somatic cells in human fetal ovaries.
An early study evaluated the number of primordial germ cells
in two human embryos in week 4 p.c. to be ∼450 and ∼1400
per embryo (Witschi, 1948). The classical study by Baker has
been the only one to estimate the number of germ cells in
human fetal sex-differentiated ovaries (Baker, 1963); the study
covered the main part of fetal life and was based on ≤14 ovaries, two from week 9 and the rest distributed among older
fetuses. Baker found a mean of 300 000 oogonia per ovary in
the two 9 week old ovaries. The range of the number of oogonia was not given. The total germinal tissue occupied 19.3 and
19.7% of the total ovarian volume respectively, but the volume
of the two ovaries cannot be deduced from the paper and the
range of the number of oogonia is thus unknown. However, the
number of 253 600 oogonia estimated in the present study from
the fetus in week 9 is comparable with the number described
by Baker. The discrepancy in the total number of oogonia
described in the two studies (15%) may lie within the physiological range, or may relate to the assessment methods of age
of the fetus and/or to the methods for estimation of the germ
cell number. Baker used the crown-rump length and the
present study applied the length of limbs and foot
(Evtouchenko et al., 1996), the latter presumably being more
accurate than the crown–rump length only. Further, the counting techniques differed profoundly. While Baker’s method
relied on volumetric calculations of oogonia and ovaries
(Chalkley, 1943) the present study used the optical fractionator
technique, a stereological method, known to provide precise
and unbiased estimates of cell numbers in different organs
(Gundersen et al., 1988).
Baker estimated the ratio between volume (not number) of
germ cells and somatic cells to be 1:5 in the two ovaries from
week 9. In the present study, the ratio between the number of
oogonia and somatic cells showed a mean of ∼1:8. Since the
volume of somatic cells is smaller than that of oogonia the
number ratio is lower than volume ratio, as would be
expected, and our estimated ratio may thus approach Baker’s
ratio if volume is taken into account. The tendency of an
increasing ratio between number of oogonia and somatic cells,
33
E.Bendsen et al.
as we report, indicates that oogonia proliferate relatively
faster than somatic cells as the first trimester fetus grows
from week 6 to week 9.
Recently we have shown that the number of germ cells in
human fetal testes of the same age is significantly smaller, the
number of prespermatogonia being ∼3000 in week 6 and
∼30 000 in week 9 p.c. (Bendsen et al., 2003). This corresponds
to a prespermatogonia doubling time of ∼6 days during this
3 week period. The present study revealed that 26 000 oogonia
became 250 000 in almost the same 3 weeks, resulting in a similar doubling time of ∼6 days. Provided the testis and the ovary
contained the same number of germ cells at the time of sex differentiation—perhaps a little less than 3000—the oogonia must
have grown three times faster than the prespermatogonia in the
beginning of the period. However, it is not known whether the
developing ovary and testis are equipped with the same number
of germ cells before the time of sex differentiation.
The testis at week 6, which contained 3000 prespermatogonia,
was barely morphologically recognizable as a testis (Bendsen
et al., 2003), so sex differentiation presumably took place
shortly before this fetus was aborted, perhaps late in week 5
p.c., which may be a little earlier than stated in the literature,
namely during week 6 p.c. (Witschi, 1963; Wartenberg, 1982).
According to previous (Bendsen et al., 2003) and present
results it can be calculated that the doubling time of somatic
cells is roughly similar in testis and in ovaries from first trimester: a little less than 6 days in ovaries (165 000 somatic cells in
the ovary from week 6 become 1 500 000 in week 9) and similar in testes (526 000 somatic cells in the testis from week 7
become 1 735 000 in week 9). At the present time it is not
known what is controlling proliferation of either germ cells or
somatic cells in early differentiating human gonads. However,
it seems unlikely that gonadal steroids are directly involved
considering the vast difference between steroid content in the
testes and ovaries at that time of development and the similarity of doubling time for both germ cells and somatic cells of the
two sexes.
Oocytes in preleptotene and leptotene were present already
in ovaries from week 9, which is 1–2 weeks earlier than we
previously reported (Gondos et al., 1986). This discrepancy
may relate to the fact that in the present study the entire ovary
is serially sectioned and all sections carefully examined. In
contrast, in all other studies only a fraction of the germ cells
has been evaluated. The finding that meiotic cells were only
seen in the cranial end of the ovary supports the suggestion that
mesonephros exerts stimulatory effects on initiation of meiosis.
In conclusion, we found that from week 6 to week 9 the
number of oogonia increased from ∼26 000 to ∼250 000 and
somatic cells from ∼240 000 to ∼1 400 000. Meiotic germ cells
were seen in the cranial end of ovaries from week 9.
Acknowledgements
The excellent technical assistance of Inga Husum, Laboratory of
Reproductive Biology, University Hospital of Copenhagen, is greatly
acknowledged. Christian Olesen and Nanna D.Rendorff, Department
of Medical Genetics, Institute of Medical Biochemistry and Genetics,
University of Copenhagen, are acknowledged for their help with sex
determination of the ovaries by PCR technique. The study was
34
sponsored by the Faculty of Health Sciences, University of Southern
Denmark, by the Danish Medical Research Council, no. 9700832 and
9502022, and by the Danish Environmental Research program
(Center for Environmental Estrogen Research).
References
Baker TG (1963) A quantitative and cytological study of germ cells in human
ovaries. Proc R Soc Lond Ser B Biol Sci 158,417–433.
Bendsen E, Laursen SB, Olesen C, Westergaard LG, Yding Andersen C and
Byskov AG (2001) Effect of 4-octylphenol on germ cell number in cultured
human fetal gonads. Hum Reprod 16,236–243.
Bendsen E, Byskov AG, Laursen SB, Larsen HP, Yding Andersen C and Westergaard LG (2003) Number of germ cells and somatic cells in human fetal testes during the first weeks after sex differentiation. Hum Reprod 18,13–18.
Byskov AG (1975) The role of the rete ovarii in meiosis and follicle formation
in the cat, mink and ferret. J Reprod Fertil 45,201–209.
Byskov AG and Hoyer PE (1994) Embrylogy of mammalian gonads and ducts.
In Knobil E and Neill JD (eds), The Physiology of Reproduction. Raven
Press, New York, pp 487–539.
Chalkley HW (1943) Method for the quantitative morphologic analysis of tissues. J Natl Cancer Inst 4,47–53.
Evtouchenko L, Studer L, Spenger C, Dreher E and Seiler RW (1996) A mathematical model for the estimation of human embryonic and fetal age. Cell
Transplant 5,453–464.
Feinstein RE, Westergren E, Bucht E, Sjoberg HE and Grimelius L (1996)
Estimation of the C-cell numbers in rat thyroid glands using the optical fractionator. J Histochem Cytochem 44,997–1003.
Gondos B, Westergaard L and Byskov AG (1986) Initiation of oogenesis in the
human fetal ovary: ultrastructural and squash preparation study. Am J
Obstet Gynecol 155,189–195.
Gundersen HJ (1986) Stereology of arbitrary particles. A review of unbiased
number and size estimators and the presentation of some new ones, in memory of William R. Thompson. J Microsc 143(Pt 1),3–45.
Gundersen HJ and Jensen EB (1987) The efficiency of systematic sampling in
stereology and its prediction. J Microsc 147(Pt 3),229–263.
Gundersen HJ, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N,
Moller A, Nielsen K, Nyengaard JR and Pakkenberg B (1988) The new stereological tools: disector, fractionator, nucleator and point sampled intercepts
and their use in pathological research and diagnosis. APMIS 96,857–881.
Hirshfield AN (1991) Development of follicles in the mammalian ovary. Int
Rev Cytol 124,43–101.
Josso N, Lamarre I, Picard JY, Berta P, Davies N, Morichon N, Peschanski M
and Jeny R (1993) Anti-mullerian hormone in early human development.
Early Hum Dev 33,91–99.
Koopman P, Munsterberg A, Capel B, Vivian N and Lovell-Badge R (1990)
Expression of a candidate sex-determining gene during mouse testis differentiation. Nature 348,450–452.
Lovell-Badge R and Robertson E (1990) XY female mice resulting from a heritable mutation in the primary testis-determining gene, Tdy. Development
109,635–646.
McKay DG, Hertig AT, Adams EC and Danziger S (1953) Histochemical
observations on the germ cells of human embryos. Anat Rec 117,201–219.
Nakahori Y, Hamano K, Iwaya M and Nakagome Y (1991) Sex identification
by polymerase chain reaction using X-Y homologous primer. Am J Med
Genet 39,472–473.
Nauert GM and Freeman TB (1994) Low-pressure aspiration abortion for
obtaining embryonic and early gestational fetal tissue for research purposes.
Cell Transplant 3,147–151.
Sterio DC (1984) The unbiased estimation of number and sizes of arbitrary
particles using the disector. J Microsc 134(Pt 2),127–136.
Wartenberg H (1982) Development of the early human ovary and role of the
mesonephros in the differentiation of the cortex. Anat Embryol (Berl)
165,253–280.
West MJ (1993) New stereological methods for counting neurons. Neurobiol
Aging 14,275–285.
West MJ and Gundersen HJ (1990) Unbiased stereological estimation of the
number of neurons in the human hippocampus. J Comp Neurol 296,1–22.
West MJ, Slomianka L and Gundersen HJ (1991) Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator. Anat Rec 231,482–497.
West MJ, Ostergaard K, Andreassen OA and Finsen B (1996) Estimation of
the number of somatostatin neurons in the striatum: an in situ hybridization
study using the optical fractionator method. J Comp Neurol 370,11–22.
Germ and somatic cells in human fetal ovaries
Winter JSD, Faiman C and Reyes FI (1977) Sex steroid production by the
human fetus: its role in morphogenesis and control by gonadotrophins. In
Blandau RJ and Bergma D (eds) Morphogenesis and Malformation of the
Genital System. Alan R Liss, New York, pp 41–58.
Witschi E (1948) Migration of the germ cells of human embryos from the yolk
sac to the primitive gonadal folds. Contrib Embrol 32,67–80.
Witschi E (1963) Embryology of the ovary. In Grady HG and Smith DE (eds),
The Ovary. The Williams & Wilkins, Baltimore, pp 1–10.
Zuckermann SS (1962) The Ovary. Academic Press, New York.
Submitted on 17 May 2005; revised on 22 June 2005; accepted on 04 July 2005
35