General and Comparative Endocrinology 148 (2006) 95–101

www.elsevier.com/locate/ygcen

IdentiWcation of a functional corpus luteum in the Atlantic hagWsh, Myxine glutinosa

Mickie L. Powell1, Scott Kavanaugh, Stacia A. Sower ¤

Department of Biochemistry and Molecular Biology, University of New Hampshire, College of Life Sciences and Agriculture, 46 College Road, Durham, NH 03824-2617, USA

Received 18 August 2005; revised 30 December 2005; accepted 2 January 2006Available online 17 February 2006

Abstract

HagWsh represent the oldest extant craniates and are an important link between invertebrates and vertebrates. However, key elementsof the reproductive system have not been elucidated in hagWsh. There is new evidence from our recent studies that Atlantic hagWsh mayhave a seasonal reproductive cycle. These data include seasonal changes in gonadotropin-releasing hormone, gonadal steroids, estradiol,and progesterone, corresponding to gonadal reproductive stages along with the putative identiWcation of a functional corpus luteum. Thecorpus luteum in non-mammalian vertebrates secretes mainly progesterone thought to be involved in the retention of eggs and down reg-ulation of vitellogenin synthesis. The most ancient vertebrate that is known to have a functional corpus luteum is the dogWsh, Squalusacanthias. However, brown bodies, hypothesized to be corpora lutea, have been observed by scientists for over 100 years in the gonad ofthe hagWsh. To date, data in support of these brown bodies acting as corpora lutea have consisted mainly of observational studies. There-fore, we examined the putative corpora lutea (post-ovulatory follicles) in hagWsh by histology, electron microscopy, and production ofprogesterone and estradiol. Progesterone concentrations from post-ovulatory follicles were signiWcantly higher (12§ 1.5 pg/mg gonad tis-sue wet weight) compared to controls containing only gonadal tissues and oocytes (3.6§ 1.5 pg/mg gonad tissue wet weight) (p < 0.05).Estradiol was detected in seven of the 13 samples containing only gonadal tissue with oocytes and ranged between 0.6 and 0.18 pg/mggonad tissue wet weight and was not detected in any of the media containing only corpora lutea samples. Light and electron microscopyanalysis supported that these structures were corpora lutea like structures (post-ovulatory follicles). From these results, we hypothesizethat hagWsh have functional corpora lutea like structures that produce progesterone.© 2006 Elsevier Inc. All rights reserved.

Keywords: Agnathan; HagWsh; Myxine glutinosa; Corpus lutea; Progesterone

1. Introduction

The corpus luteum is a transient endocrine gland thatforms in the ovary of mammals and some non-mammalsfrom the follicular layers after ovulation. The evolution ofthe corpus luteum is associated with the event of viviparityin mammals and is necessary for the production of proges-terone that is necessary for growth and maintenance of

* Corresponding author. Fax: +1 603 862 4013.E-mail address: sasower@cisunix.unh.edu (S.A. Sower).

1 Present address: Department of Biology, University of Alabama atBirmingham, Birmingham, AL 35294-1170, USA.

0016-6480/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.ygcen.2006.01.003

pregnancy (Juengel et al., 1999). In non-mammalian verte-brates, the corpus luteum forms from the post-ovulatoryfollicles or from atretic follicles and secretes mainly proges-terone thought to be involved in the retention of eggs andregulation of hepatic vitellogenin synthesis (ChieY andChieY-Baccari, 1999). The corpus luteum shares character-istics with other steroid producing cells, such as extensiveendoplasmic reticulum, large numbers of mitochondria andthe accumulation of lipid used in the synthesis of steroidhormones.

The corpus luteum has been identiWed in birds, Wsh, rep-tiles, and amphibians (ChieY and ChieY-Baccari, 1999)(Fig. 1). The oldest extant jawed vertebrate in which the

96 M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101

corpus luteum had been positively identiWed is the dogWsh(Squalus acanthias) (Tsang and Callard, 1987). Earlier stud-ies of Hisaw and Hisaw (1959) reported the formation ofthe corpora lutea in six species of elasmobranch Wshes. Todate the corpus luteum of pregnancy in the dogWsh is phy-logenetically the most ancient and ontogenetically the lon-gest lived functional structure of this kind known in thevertebrate kingdom (Tsang and Callard, 1987). The forma-tion of a true corpus luteum in the most ancient vertebrates,the cyclostomes, has not been resolved. The presence of aluteal like structure in cyclostomes has been observed butits function has remained an enigma. Cyclostomes are com-posed of the Petromyzonitidae (lampreys) and the Myxini-dae (hagWsh). Lampreys spawn only once in their life andthen die, hence there is no further development of the struc-tures remaining in the ovary after ovulation (Saidapur,1982). HagWsh however are thought to be multiple spaw-ners and several researchers have described “brown bodies”in the hagWsh gonad that appear to be post-ovulatory folli-cles or atretic eggs in the process of resorption (Powellet al., 2005; Kosmath et al., 1983A,B; Lyngnes, cited inWalvig, 1963).

The adult hagWsh ovary is comprised of germinal tissuethat is restricted to the free edge of the membranousgonad (Sower and Gorbman, 1999). Female hagWsh pro-duce between 20 and 30 yolky, shelled eggs at a time.Numerous small oocytes move upward within the mem-brane as they grow. As the development proceeds, theeggs change shape from round to oblong, and Wnally to aspindle shape (Sower and Gorbman, 1999). Females donot have oviducts so the mature eggs are released into theabdominal cavity (Gorbman, 1997). The eggs vary in sizefrom 14 to 25 mm, with the size depending mostly uponthe species (Patzner, 1978). The eggs usually have hookedWlaments at either end that keep the eggs connectedtogether. This enables them to be ejected in a long chain.A current theory is that the eggs are ovulated through thecloacal pore and are expelled under pressure by muscularcontractions (Gorbman, 1997). Empty follicles thatappear as brown nodes remain in the ovary after the egghas been discharged.

Fig. 1. Schematic diagram of the presence or absence of corpora lutea invertebrates.

The histology of the ovary as described by Lyngnes(1936, cited in Walvig, 1963) includes two major group-ings beyond the normal developing eggs referred to asatresia or degeneration of non-ovulated follicles andregression of ovulated follicles. The post-ovulatory folli-cles in Myxine are of two kinds, solid tissue masses orXuid Wlled cysts (Walvig, 1963). These structures formthrough a process of follicular atresia accompanied byhypertrophy of the follicular cells. Degeneration of theovulated follicles is believed to give rise to the yellow orbrown bodies resembling corpora lutea in the gonads ofhagWsh. There have been several detailed studies of theatretic follicles and regression of the ovulated follicles(reviewed in Walvig, 1963). Resorption of the regressionof ovulated follicles appears to occur slowly since it ispossible to distinguish as many as four generations of fol-licles (corpora lutea) in various degrees of involution(Walvig, 1963). However, the function of these folliclesremains unclear. Tsuneki and Gorbman (1977) and Kos-math et al. (1983A,B) performed electron microscopicaland histochemical investigations in post-ovulatory folli-cles in Eptatretus stouti and M. glutinosa, respectively,and did not Wnd any evidence for hormone-producingcells. However, ChieY and Botte (1970) demonstrated thepresence of cholesterol in the post-ovulatory follicles,which is associated with the luteinization process and theformation of the corpus luteum. Progesterone has beenidentiWed in the gonads of the hagWsh species Eptatretusburgeri and Myxine glutinosa (Hirose et al., 1975; Powellet al., 2004) but little is known about the steroids associ-ated with these brown bodies and the steroid productionof the brown bodies alone has not been examined in hag-Wsh.

To test the hypothesis that the brown/yellow bodies(post-ovulatory follicles) in the hagWsh gonad function ascorpora lutea, progesterone, and estradiol concentrationsof the brown bodies and gonadal tissue containing onlyoocytes were determined. The tissues were also examinedhistologically using both light and electron microscopy.

2. Materials and methods

2.1. HagWsh collection

Atlantic hagWsh M. glutinosa, identiWed according to the characteris-tics and geographic locations described by Bigelow and Schroeder (1948),were trapped monthly from November 2001, to October 2002, at JeVeriesLedge (42°50.6�N, 70°10.079�W) in the Gulf of Maine using the Universityof New Hampshire Research Vessel Gulf Challenger as part of an ongoingstudy on growth and reproduction. HagWsh were trapped using modiWed200 L plastic barrels baited with salted herring and Wtted with one-waycones. Traps were set for 45–60 min at a depth of 100–150 m. Large hagWsh45–55 cm+ from the October 2002, sample were chosen to increase theprobability of obtaining mature female hagWsh with brown bodies. Hag-Wsh were maintained in 4 °C seawater and transported to the University ofNew Hampshire, Durham, New Hampshire and held for 24–48 h in chilledseawater (4 °C) tanks until dissected.

The Atlantic hagWsh used for electron microscopy were caught oV thecoast of New England by New England Eel Processors (Gloucester, MA)during February and March of 2005.

M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101 97

2.2. Tissue collection for radioimmunoassay and microscopy

HagWsh were anesthetized by immersing in a solution of 0.05 g/L trica-nine methylsulfonate (MS222), weighed, measured for length, and decapi-tated. Thirteen large hagWsh containing brown bodies in their gonads wereidentiWed during dissections. Tissue samples containing only brown or yel-low bodies (representing post-ovulated follicles) or only gonad tissue withdeveloping oocytes were then removed. Post-ovulatory follicles (early,middle, and late stages of luteal regression) were selected for electronmicroscopy studies based on the size and color in which the early stageswere yellow and larger and as regression progressed, the follicles becamesmaller and the color changed to brown.

2.3. Microscopy

The tissues were placed in individual wells of a 24-well culture platewith 500 �L buVered saline (Hirose et al., 1975). The tissues were preincu-bated for one hour at 4 °C on a shaker. The preincubation media wereremoved and replaced with 500 �L pregnenolone in buVered saline(150 ng/mL media), the tissue was incubated with shaking for 48 h at 4 °C.After 48 h the tissues were removed, blotted and weighed. Tissues wereWxed in Bouin’s solution, embedded in wax, sectioned (7 �m thickness) andstained with eosin and hematoxylin for light microscopy. Isolated corporalutea were Wxed in modiWed Karnovsky’s Wxative (2% gluteraldehyde, 2%formalin, and 0.1 M Na cacodylate buVer, pH 7.2) then rinsed with caco-dylate buVer, pH 7.2, and post-Wxed with 1% OsO4 in 0.1 M Na cacodylatebuVer, pH 7.2 (Karnovsky, 1965). The samples were then rinsed with caco-dylate buVer, pH 7.2, followed by dehydration and embedded using Poly/Bed 812 Embedding Media/DMP-30 Kit (Polysciences #08792-1) using astandard Mollenhauer mixture (Mollenhauer, 1964). Thick sections (1 �m)were taken and stained with toluidine blue for light microscopy for orien-tation purposes. Selected samples were then thin sectioned, placed on cop-per grids, and stained with lead citrate and uranyl acetate. The sampleswere photographed with a Leo/Zeiss 922 omega transmission electronmicroscope.

2.4. Radioimmunoassay

The culture media were collected and frozen at ¡80 °C untilextracted, and assayed for estradiol and progesterone by radioimmuno-assay (RIA) following the procedures described by Sower and Schreck(1982). The RIA for estradiol-17� used antiestradiol-17� (S-244)obtained from G. Niswender (Colorado State University, Fort Collins).The estradiol antiserum was used at a dilution of 1:85,000. The lowerlimit of detection was 0.488 pg/0.1 mL. The intraassay and interassaycoeYcients of variation for the estradiol RIA were 5.4 and 6%, respec-tively. The RIA for progesterone used antiprogesterone (337) obtainedfrom G. Niswender (Colorado State University, Fort Collins). The pro-gesterone antiserum was used at a dilution of 1:8500. Cross-reactivitiesof progesterone antiserum 337 were shown to be 100% progesterone;3.7% pregnenolone; 0.8% testosterone, and n.d. estradiol (Koligian andStormshak, 1977). The lower limit of detection was 7.8 pg/0.1 mL. Theintraassay and interassay coeYcients of variation for the progesteroneRIA were 2 and 5%, respectively.

2.5. Statistics

SigniWcance was calculated using Student’s t test (n D 13, p < 0.5).

3. Results

3.1. Histology

3.1.1. Light microscopyThe brown/yellow bodies (post-ovulatory follicles) were

present in the gonad tissue of hagWsh with maturing eggs.

The brown/yellow bodies “corpora lutea” were of two typeseither circular, solid, Wlled with cellular material or elon-gated with little material inside the structure. Solid “cor-pora lutea” were Wlled cells that were less densely packedthan those occurring on the outer margins of the “corpusluteum.” Both types of corpora lutea had well deWned lay-ers of cells around the outer margin (Fig. 2). This well-deWned layer of cells is arranged concentrically around thefollicle and resembles the theca lutein cells surrounding a“corpus luteum.” The inner layer of cells is much thickerand resembles the granulose lutein cells of the mature “cor-pus luteum.” The gonad tissue contained oocytes of varioussizes but no brown bodies “corpora lutea” (Fig. 3). Theouter cell layer is well deWned in the oocytes.

3.1.2. Electron microscopyExtensive endoplasmic reticulum was observed in the

brown/yellow bodies (post-ovulatory follicles). Large num-bers of mitochondria and lipid droplets were also present.Examination of the post-ovulatory follicles revealed vari-ous stages of resorption (progression of luteolysis) wereobserved. In the early stages of luteal regression early stagesof nuclear breakdown (N), nucleus with fragmented massesof chromatin (C), distended smooth endoplasmic reticulum(SER), particulate bodies (P), and lipid accumulations (L),were observed in the regressing brown bodies (Fig. 4). Mid-Stage of luteal regression was characterized by a continua-tion of organelle decay an increase in lipid droplets and anincrease in size of Wbroblast (especially attenuations). Therewas also an increase in the amount of connective tissueobserved (Fig. 5). Late stages of luteolysis were identiWedby complete breakdown of the nucleus and distension oforganelles (Fig. 6).

3.2. Steroids

Previous experiments measuring the progesterone con-centration in the gonad tissues of M. glutinosa have shownprogesterone is present in the gonad tissue of hagWsh rang-ing in size from 18 to 65 cm sampled from April 2001 toOctober 2002, in concentrations between 10 and 2000 pg/mg gonad tissue wet weight (Powell et al., 2004). In thisstudy, the progesterone concentrations ranged between 2and 56 pg/mg gonad tissue wet weight and were detected inall brown/yellow bodies (post-ovulatory follicles) andgonadal tissues. In the present study, the concentrations ofprogesterone in media containing post-ovulatory follicleswere signiWcantly (p < 0.05) higher than media containingonly oocytes (Fig. 4). The average concentration of proges-terone measured in female corpus luteum was 12§ 1.5 pg/mg gonad tissue wet weight (see Fig. 7). The lowest concen-tration of progesterone (1.9 pg/mg gonadal tissue wetweight) was measured in follicles that had recently releasedeggs and had not begun to reabsorb. Estradiol was detectedin seven of the 13 samples containing only gonadal tissuewith oocytes and ranged between 0.6 and 0.18 pg/mg gonadtissue wet weight.

98 M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101

Fig. 2. HagWsh gonad with maturing eggs and corpora lutea. Arrows indicate corpora lutea. (A–C) Corpora Lutea of hagWsh. Arrows indicating solid cor-pora lutea from the gonad of M. glutinosa. (D) Arrow indicating corpus luteum in later stages of resorption, structures have a more narrow elongatedform (all views 40£).

M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101 99

Fig. 3. Representative section of gonadal tissue lacking corpora lutea.Various size oocytes are indicated by arrows (40£).

4. Discussion

In the current study, we have shown that hagWsh have cor-pora lutea like-structures (post-ovulatory follicles) that pro-duce progesterone. This is the Wrst reported study on possiblefunctional corpora lutea in hagWsh, one of two extant repre-sentatives of the oldest lineage of vertebrates. Conel (1917)interpreted the identiWed brown bodies as corpora luteaformed from reabsorbing follicles and atretic eggs. Lyngnes(1936, cited in Walvig, 1963) reported that well-developedcorpora lutea are formed in both the atretic and ruptured fol-licles of Myxine, but there was nothing known concerningtheir functional activity. All of the hagWsh examined in thecurrent study contained vitellogenic eggs ranging in size from8 to 25mm and “corpus luteum.” The histology of the “cor-pus luteum” was similar to that described by Lyngnes (1936,cited in Walvig, 1963) which includes two major groupingsbeyond the normal developing eggs referred to as atresia ordegeneration of non-ovulated follicles and regression of ovu-lated follicles. In the current study, various ovulated folliclesand atretic follicles were observed and could be distinguished

Fig. 4. Early stages of luteal regression. (A) Low magniWcation of transforming luteal and theca cells. (B–D) Distended smooth endoplasmic reticulum(SER), lipid accumulations (L), mitochondria with tubular cristae (M) and junction between two luteal cells (LJ) (A, 2000£; B, 12,000£; C, 5000£; D,

M

SERSER

M

LJ

L

A B

DC

30,000£).

100 M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101

Fig. 6. Late stages of luteolysis were marked by complete breakdown ofthe nucleus and distension of organelles, Wbroblast (F), degraded nucleusof luteal cell (LC), connective tissue (CT) cell (10,000£).

Fig. 7. Concentration of progesterone measured by radioimmunoassay inthe corpora lutea and the gonad tissue of M. glutinosa. SigniWcant diVer-ence between the groups (*p < 0.05).

microscopically and by their color as observed by otherresearchers (cited in Walvig, 1963). Generally, it is only thepost-ovulatory follicles that are often yellow or brown incolor and this is due to the contracting and changes of thefollicle into a long, thin bag or Xat bud-shaped formationwith intermediate stages along with the shrinkage of the fol-licular ligament contracts as deWned by Lynges (cited in Wal-vig, 1963). The electron micrographs provided the mostdetailed view of the hagWsh post-ovulatory follicle to dateand veriWes the presence of structures known to be necessaryfor steroid synthesis. These included extensive endoplasmicreticulum, numerous mitochondria and lipid droplets. As thecorpus luteum was resorbed, nuclear degeneration, fragmen-tation of chromatin and distended smooth endoplasmic retic-ulum became evident. The process of resorption appears tobe quite slow as some hagWsh may have as many as four gen-erations of post-ovulatory follicles at diVerent stages ofresorption (Dodd and Dodd, 1985). In the present study, theconcentration of progesterone measured in the “corpus lut-eum” was signiWcantly higher than that in the surroundingovary containing only oocytes. Estradiol was not detected inthe incubation media of the “corpus luteum” suggesting thatestradiol may not be produced.

In other species of vertebrates, the corpus luteum formsafter ovulation as the granulosa cells begin to accumulatelarge quantities of cholesterol necessary for the production ofsteroids such as progesterone. In mammals, one of the threemajor functions of progesterone is to maintain the uterus in ahighly vascularized state; in non-mammalian species the cor-pus luteum is considered to be involved in vascularization ofgonad tissues and the retention of eggs (ChieY and ChieY-Baccari, 1999). Small quantities of cholesterol have beendemonstrated by the Schultz test in the post-ovulatory folli-cles of hagWsh (Saidapur, 1982). The present study identiWesby electron microscopy an increase in lipid droplets whichmay correspond to the increased progesterone productionassociated with the hagWsh corpus luteum (brown bodies).Conel (1931) proposed that the growth of the eggs in hagWshis dependent on the blood supplied to each large egg by a sin-gle ovarian artery. In the post-ovulatory follicles, both thenucleus and cell components degenerate and form an epithe-

Fig. 5. Mid-Stage of luteal regression was characterized by a continuation of organelle decay an increase in lipid droplets and an increase in size of Wbro-blast (especially attenuations) particulate bodies (P), lipid accumulations (L), connective tissue (CT), distended smooth endoplasmic reticulum (SER),nucleus with fragmented masses of chromatin (C) and nuclear breakdown (N) (A, 12,500£; B, 5000£).

P

L

N

C

A B

SER

CT

M.L. Powell et al. / General and Comparative Endocrinology 148 (2006) 95–101 101

lial mass; blood vessels form in this epithelia mass in thesame way as in later evolved vertebrates (reviewed in Walvig,1963). Although the function of progesterone in hagWshremains unknown, we propose that this growth of vasculari-zation in the hagWsh gonad may be due to progesterone pro-duced by the corpus luteum (post-ovulatory follicles). Thiswill require further investigation to determine if this hor-mone action has been conserved throughout vertebrates.

In hagWsh only 25–30 eggs (average length of 20 mm)mature at one time, the remaining eggs will arrest their devel-opment at 2–5 mm in length and remain in this arrested stateuntil the selected eggs have matured and been released. Afterthe eggs are released the reabsorbing egg follicles appear toproduce the progesterone necessary for the vascularizationand maturation of the next series of eggs. A correlationbetween the resorption of the follicles and steroid concentra-tion in the gonad incubation media was observed in at leastone of the hagWsh examined in this study. The lowest concen-trations of progesterone and estradiol were measured in afemale that had recently released eggs and had not begun theprocess of resorption of the follicle, suggesting that steroidproduction in the hagWsh corpus luteum may be related tothe degree of follicle resorption. Dodd and Dodd (1985) pro-posed that a reciprocal action of pre- and post-ovulation cor-pora lutea (brown/yellow bodies) hormones regulate thedevelopment of a series of eggs, as well as morphological andbehavioral sex characters prior to spawning. HagWsh as basalextant vertebrates have the lowest arterial pressure of anyvertebrate species (Forster, 1998) and may have requiredhormone mediated action for vascularization. Further inves-tigations will be required to determine if progesteronereleased from post-ovulatory follicles is involved in vascular-ization and is thus representative of an ancestral functionthat anteceded vertebrate evolution.

Acknowledgments

We thank Cari Bourn, Mihael Freamat, NathanielNucci, Adam Root, Matt Silver, JoAnne Davis, Jen Glieco,Emily Violette, Jocelyn Sanford, and Greg Blaisdell, forassistance in collection of animals and processing of sam-ples. We thank Yang Cho of New England Eel Process,Gloucester Massachusetts for providing hagWsh, and Dr.Mark Townley and Nancy Cherim for their assistance withsample preparation and technical help with electronmicroscopy. We thank the crew of the RV Gulf Challenger,Captain-Paul Pelletier, Bryan Soares, and Debbie Brewittfor the use of the boat, and other assistance. This researchwas supported by UNH/UME NOAA Sea Grant R/FMS-168. This is ScientiWc Contribution Number 2227 from theNew Hampshire Agricultural Experiment Station.

References

Bigelow, H.B., Schroeder, W.C., 1948. Cyclostomes. In: Parr, A.E. (Ed.),Fishes of the Western North Atlantic. Sears foundation for MarineResearch, Yale University, New Haven, pp. 34–38.

ChieY, G., Botte, B., 1970. The problem of “luteogenesis” in non-mamma-lian vertebrates. Boll. Zool. 37, 85–102.

ChieY, G., ChieY-Baccari, G., 1999. Corpora lutea of nonmammalian spe-cies. In: Knobil, E., Neill, J.D. (Eds.), Encyclopedia of Reproduction.Academic Press, San Diego, pp. 680–688.

Conel, J.R., 1917. The urogenital system of Myxinoids. J. Morphol. 29,75–163.

Conel, J.L., 1931. In: Gudger, E.W. (Ed.), The Bashford Dean MemorialVolume: Archaic Fishes. American Museum of Natural History, NewYork, pp. 64–102.

Dodd, J.M., Dodd, M.H.I., 1985. Evolutionary aspects of reproduction incyclostomes and cartilaginous Wshes. In: Formen, R.E., Gorbman, A.,Dodd, J.M., Olsson, R. (Eds.), Evolutionary Biology of PrimitiveFishes. Plenum Press, New York, pp. 295–319.

Forster, M.E., 1998. Cardiovascular function in hagWshes. In: Jorgensen,J.M., Lomholt, J.P., Weber, R.E., Malte, H. (Eds.), The Biology ofHagWshes. Chapman and Hall, London, pp. 237–259.

Gorbman, A., 1997. HagWsh development. Zool. Sci. 14, 375–379.Hirose, K., Tamaoki, B., Fernholm, B., Kobayashi, H., 1975. In vitro bio-

conversions of steroids in the mature ovary of the hagWsh, Eptatretusburgeri. Comp. Biochem. Physiol. B 51, 403–408.

Hisaw Jr., F.L., Hisaw, F.L., 1959. Copora lutea of elasmobranch Wshes.Anat. Rec. 135, 269–277.

Juengel, J.L., McIntush, E.W., Niswender, G.D., 1999. Corpus luteum. In:Knobil, E., Neill, J.D. (Eds.), Encyclopedia of Reproduction. AcademicPress, San Diego, pp. 703–709.

Karnovsky, M.J., 1965. A formaldehyde–glutaraldehyde Wxative of highosmolarity for use in electron microscopy. J. Cell Biol. 27, 137A.

Koligian, K.B., Stormshak, F., 1977. Nuclear and cytoplasmic estrogenreceptors in ovine endometrium during the estrous cycle. Endocrinol-ogy 101 (2), 524–533.

Kosmath, I., Patzner, R.A., Adam, H., 1983A. Regression in the ovary ofMyxine glutinosa L. (Cyclostomata) II. Electron microscope studies ofatretic follicles. Z. Mikrosk Anat. Forsch. 97, 667–674.

Kosmath, I., Patzner, R.A., Adam, H., 1983B. Cyclical changes in theovary of the hagWsh Eptatretus burgeri (Cyclostomata). Z. MikroskAnat. Forsch. 97, 941–947.

Lyngnes, R., 1936. Ruckbildung der ovulierten und nicht ovulierten folli-kel im ovarium der Myxine glutinosa L. Norske videnskaps-akademia ioslo. Matematisk-naturvidenskapelig klasse 4, 1–116. (cited in Walvig,F., 1963. The gonads and the formation of the sexual cells. In: Brodal,A., Fange, R. (Eds.) The Biology of Myxine. Scandinavian UniversityBooks, Oslo, Norway, pp. 530–580).

Mollenhauer, H.H., 1964. Plastic embedding mixtures for use in electronmicroscopy. Stain Technol. 39, 111–114.

Patzner, R.A., 1978. Cyclical changes in the ovary of the hagWsh Eptatretusburgeri (Cyclostomata). Acta Zool. (Stockh.) 59, 57–61.

Powell, M.L., Kavanaugh, S.I., Sower, S.A., 2004. Seasonal concentrationsof reproductive steroids in the gonads of the Atlantic hagWsh, Myxineglutinosa. J. Exp. Zool. 301, 352–360.

Powell, M.L., Kavanaugh, S.I., Sower, S.A., 2005. Current knowledge ofhagWsh reproduction: Implications for Wsheries management. Int.Comp. Biol. 45 (1), 158–165.

Saidapur, S.K., 1982. Structure and function of postovulatory follicles(corpora lutea) in the ovaries of nonmammalian vertebrates. Int. Rev.Cytol. 75, 243–285.

Sower, S.A., Gorbman, A., 1999. Agnatha. In: Knobil, E., Neill, J.D. (Eds.),Encyclopedia of Reproduction. Academic Press, New York, pp. 83–90.

Sower, S.A., Schreck, C.B., 1982. Steroid and thyroid hormones during sex-ual maturation of coho salmon (Oncorhynchus kisutch) in seawater offresh water. Gen. Comp. Endocrinol. 47, 42–53.

Tsang, P., Callard, I.P., 1987. Luteal progesterone production and regulationin the vivaparous dogWsh, Squalus acanthias. J. Exp. Zool. 241, 377–382.

Tsuneki, K., Gorbman, A., 1977. Ultrastructure of the ovary of the hagWshEptatretus stouti. Acta Zool. (Stockholm) 58, 27–40.

Walvig, F., 1963. The gonads and the formation of the sexual cells. In:Brodal, A., Fange, R. (Eds.), The Biology of Myxine. ScandinavianUniversity Books, Oslo, Norway, pp. 530–580.