Pacific Walrus (Odobenus rosmarus divergens)

Order: Pinnipedia

Family: Odobenidae

1) General Zoological Data

Walruses are large, gregarious animals with both sexes having tusks. Males are larger (1,500 kg) than females (900 kg) and occur in Northern parts of Atlantic and Pacific oceans. Mating is said to take place in January/February and is followed by a 4-5 months delay in implantation (Nowak, 1999). Birth occurs 10-11 thereafter, with the neonate weighing 50-63 kg. Several subspecies are nominated: Odobenus rosmarus rosmarus for the Atlantic animal and O. rosmarus divergens for the Pacific species; others are less well defined. The name (Odobenus) derives from the Greek for tooth and walking, as the animals have been observed to pull themselves over ice with their modified canines (tusks); and rosmarus derives from the Scandinavian denomination of the animal (Gotch, 1979). Longevity is 30 years according to Nishiwaki (1972). From his extensive study of pinniped chromosomes, Arnason (1974b) suggested a monophyletic origin of Pinnipedia. Different points of view were discussed by Duffield (1972). In their wide-ranging study of carnivore genomes, O'Brien et al. (1999) deduced that pinnipeds split from other carnivores approximately 30 MYA. Phylogenetic relations among the pinnipeds were strongly supported by the mitochondrial rRNA study of Ledje & Arnason (1996). Andersen et al. (1998) studied the population structure of Atlantic walruses with mtDNA fragments and found one animal with a Pacific walrus haplotype, thus proving a connection between the two populations.

Adult male Pacific walrus.

2) General Gestational Data

The total length of gestation is 15-16 months (including the period of delayed implantation) with singletons the rule. This delay of implantation is about 4-5 months long. There is one record of a set of twins, but the article was not available to me (Nikulin, 1954), and Hayssen et al. (1993) make reference to rare other twin gestations in Russian observations. Hybrids with other species are unknown. General data on pinniped reproduction, including walrus, may be found in the chapter by Costa & Crocker (1998).

3) Implantation

There are no publications of implantation, and Mossman (1987) described only one zonary placenta that he had observed superficially; he found it to be zonary.

4) General Characterization of the Placenta

The placenta of this term pregnancy was kindly made available by Dr. Judy St. Leger of Sea World, San Diego. It comes from a multigravida that had gained 800 lbs. during gestation and that was initially very protective of the newborn calf, but eventually the newborn had to be hand-raised. The placenta had remained in sea water for over a day but was cooled and thus in an excellent macroscopic condition. There was minimal autolysis, thus histology was also satisfactory but inadequate for electronmicroscopy. Ordinarily, the placenta is eaten by gulls, according to Hayssen et al. (1993).

This is a typical zonary placenta weighing 5,900 g and measuring 85 cm in greatest length. The "ring" of placental tissue was 25 cm wide. It was complete and the circle was 100 cm in circumference. The villous tissue was 2-2.3 cm thick and darkly congested. At the sides there was the characteristic, irregular, brownish old hematoma, the so-called hemophagous organ. The maternal surface had much fresh clot attached and was irregular on its yellowish surface. There were no subdivisions of the placental tissues. The fetal surface was smooth although a focus of white, nodular squamous metaplasia of amnion was found (5x2 cm). The umbilical cord was mostly absent and less than 10 cm remained.

Another placenta was kindly made available by Dr. Carlos Rodriguez of the New York Aquarium. It weighed 4,980 g, had a similar zonary shape and the same "hemophagous" peripheral region.

Fetal surface of walrus placenta with membranes that contain the remains of the umbilical cord at top right.

Maternal surface with yellow/brown marginal "hematomas"; the umbilical cord extends to top left.

Higher magnification of the zonary placenta, maternal surface.

Cross section of disk with brown marginal hematoma and congested villous portion.

5) Details of fetal/maternal barrier

The walrus placenta, like that of other pinnipeds, is labyrinthine and much resembles in its structure that of other carnivores, like the cat. The marginal "hematoma" is also brown, rather than green as in the dog. I did not find regions of active bleeding nor was there the central "hematoma" described by others for seal placentas. There are, however, different regions that can be identified grossly, but better microscopically, when inspecting the cross-sectioned organ. The largest portion is the labyrinth which, in my material, was very much congested. On top of the labyrinth of course is the chorioallantoic mesenchyme (the chorionic plate with fetal vessels). The base of the delivered placenta had some of the so-called "spongy layer" attached but it lacked most of the maternal endometrial glands that otherwise make up much of this zone. The base had debris, small bits of endometrial glands surrounded by syncytium, fibrin and fresh blood. Although decidua exists, little comes away with the delivered organ. At the base of the labyrinth there are irregular collections of syncytiotrophoblast, the much darker cells in the photographs shown below. The deep red color of the labyrinth of the walrus placenta shown here is from the presence of enormously dilated maternal blood channels. They are lined by endothelium and covered on the outside by trophoblast. This is thus an endotheliochorial organ.

The trophoblast at the base is mostly syncytial in nature and the quite inapparent cytotrophoblast in the labyrinth appears to give rise to the syncytium. The syncytium is superficially invasive into the endometrium, destroys some of the stroma and opens some glands, but it does not grow into the maternal vascular lumens or into their walls.

Margin of the placenta with congested villous tissue at right.

Section of one margin of the placenta with congested villous tissue at left and degenerated villous tissue and old fibrin clots at right.

Overview of cross section of placenta. Note the extensions of the chorionic connective tissue (with fetal vessels) into the labyrinth and the large maternal artery (left) penetrating deeply into the labyrinth. Endometrial remnants and trophoblast in the "junctional zone" at the base with much fibrinoid.

Fetal surface of the placenta with large fetal chorioallantoic vessel above.

labyrinth above (4/5th) and junctional zone (1/5th) below.

Higher magnification of villous region with fetal stem above left. The large spaces are filled with maternal blood (here mostly empty), lined by endothelium and covered by (dark) trophoblast.

The large maternal blood-filled sinusoids of the labyrinth are lined by thin endothelium; between them is the connective tissue of the fetus with capillaries.

Higher magnification of previous view of the labyrinth. Note maternal endothelium and darker trophoblast.

Base of the placenta, the so-called junctional zone with dilated maternal channels of the labyrinth (above), fibrinoid infiltrated by trophoblast (dark blue) and small remnants of decidua at the bottom.

Higher magnification of junctional zone with remnants of partially destroyed endometrial glands bordered by syncytium and fibrinoid.

6) Umbilical cord

The normal length of the umbilical cord is unknown. This specimen lacked most of its cord and one vein had already split off from its cross section near the placental surface. Normally it probably contains two arteries, two veins and a large allantoic duct. Numerous small blood vessels are also present that become more numerous near the placental surface where they spread over the allantoic sac. There were no nodules on the remains of this cord's amnionic surface. The umbilical cord of the New York specimen was between 45 and 50 cm long.

Cross section of cord immediately prior to its insertion on the placental surface - one vein has already departed. The central empty area is the beginning of the allantoic sac.

7) Uteroplacental circulation

This has not been described and could not be observed in this placenta other than the visualization of endometrial arteries entering the labyrinth. They are large channels and send blood to the undersurface of the chorion whence the blood is returned through the labyrinthine channels.

8) Extraplacental membranes

The amnion is lined by a single layer of cuboidal epithelium and has one larger area of squamous inclusions. It contains no blood vessels. The large allantois has a diffuse capillary vasculature and is lined by cuboidal to columnar epithelium. Yolk sac remnants were not found in this term placenta.

The amnion over the placental surface has a single layer of cuboidal epithelium.

The amnion (above) had a region of extensive squamous metaplasia with numerous keratin inclusions. The allantoic epithelium is below.

Higher magnification of the squamous keratin balls of the amnion.

9) Trophoblast external to barrier

The trophoblastic invasion of the decidua is very superficial and, in the absence of a uterus with implanted placenta, no statement can be made about deeper trophoblast invasion. This is absent, however, in other pinnipeds and, since the larger endometrial vessels are not invaded (see pictures above), it is unlikely that deeper penetration of trophoblast occurs in walrus.

10) Endometrium

Walruses have a bicornuate uterus. The glands of the endometrium were partially destroyed in this term placenta but their bases remained. They did not contain any obvious secretion. True decidual transformation cannot be seen in the endometrial stroma.

11) Various features

As many other carnivora, at the margin of this zonary placenta one finds the "hemophagous zone", a border composed of degenerating villi, fibrin and remnants of probable hematomas. In contrast to the dog, it is not green but brown/yellow and laminated, much like that seen in the cat, as was also remarked by Amoroso (1961) for related pinnipeds. Yellow pigment occurs here and a few other areas of the placenta which has frequently a crystalline form. The crystals are highly birefractory in polarized light (unlike hemosiderin) and have a box-like shape. They do not stain for iron with Prussian blue stains. Beneath the chorion the more columnar trophoblast contains many crystals.

Composite of several polarized photographs of the marginal "hemophagous zone".

Some of the yellow crystals found in debris and subchorionic trophoblast at the edge of the zonary placenta. (Polarized light).

12) Endocrinology

I am not aware that any publications on the endocrine performance of pregnant walruses exist. A general survey of the reproductive physiology, including some general aspects of endocrine performance of pinnipeds, is found in Costa & Crocker (1998).

13) Genetics

Pinnipedia have either 32 or 34 chromosomes; the Walruses have 32 chromosomes with one metacentric pair having typical secondary constrictions (Fay et al., 1967; Arnason, 1974a; Hsu & Benirschke, 1977; Arnason, 1977; Arnason, 1981). The monophyletic origin of pinnipeds was affirmed by Arnason's study (1974b). The NF of pinnipeds is 64. A characteristic secondary constriction of a pair metacentrics and the very small Y chromosome are typical.

Chromosomes of male and female walruses (Courtesy U. Arnason).

14) Immunology

I know of no publications on the immunology of walruses.

15) Pathological features

Various surveys for antibodies have been conducted in walruses and other pinnipeds. Calle et al. (2002) found no antibodies against Brucella sp. in the Pacific walrus, 3 of 40 had antibodies against Leptospira interrogans, but phocine distemper was not detected. A variety of other antibodies existed (against calicivirus, vesicular exanthema, and various subtypes of influenza A). Nielsen et al. (2000) detected antibodies against morbillivirus in 50 of Atlantic walruses and, in 2001, identified antibodies to Brucella but they were unable to identify the organism. Although Zarnke et al. (1997) identified antibodies to phocid herpesviruses, they were of the opinion that herpes infection was not a cause of mortality. Antibodies against caliciviruses were detected in Pacific walruses by Barlough et al. (1986) who had also earlier identified the virus.

Trichinella infection is common in walrus and it is a source of infection to the Inuits (Proulx et al., 2002). Brucellosis of sea mammals was studied by Foster et al. (2002) who separated the responsible organism of terrestrial animals from those in sea mammals (Brucella cetaceae & B. pinnipedia). A listing of parasites found in walruses was provided by Dailey & Brownell (1972).

Myocardial infarction and widespread atherosclerosis was the cause of death in a 25 year-old walrus reported by Gruber et al. (2002).

Cornell et al. (1975) described an unusual complication of neonatal life, the persistence of a tight fibrous band expanding from the dome of the bladder to the umbilicus that prevented complete emptying of the bladder. The about one month-old, 69 kg walrus had been hand-reared and developed pyelonephritis and hydronephrosis from this complication and died. It also had ascites and intestinal torsion.

16) Physiologic data

No physiological studies on placenta or uterine perfusion have been undertaken on walrus implantations. There is, however, an excellent consideration of the development of the (probably similar) endotheliochorial placentation in mink (Pfarrer et al., 1999). It described the development of the maternal sinusoids from endometrial capillaries through the invasive action of the syncytiotrophoblast in three developmental stages. Histologic studies and corrosion casts were employed and these authors provided an excellent overview of the features of placental perfusion in endotheliochorial placentas.

17) Other resources

To the best of my knowledge, the only cell lines available are those of Dr. U. Anderson at the University of Lund, Sweden.

18) Other remarks - What additional Information is needed?

No implantational study has been undertaken and the length of the umbilical cord needs to be ascertained. Endocrine studies are completely lacking.

Acknowledgement

The animal photograph in this chapter was kindly supplied by Dr. Sam Ridgway. I very much appreciate the help of Dr. Judy St. Leger of Sea World, San Diego for allowing me to study this placenta. I also wish to acknowledge the constant help of the histopathologists at UCSD Medical Center in preparing the slides.