(Clicking on the thumbnail images below will launch a new window and a larger version of the thumbnail.)
Last updated:
March 21, 2007.
Elephas maximus & Loxodonta africana

Order: Proboscidea
Family: Elephantidae

1) General zoological data of species

There are two elephant species, the Asian elephant (Elephas maximus), and the somewhat larger African elephant (Loxodonta africana). Their evolutionary history has been studied in some detail by Aguirre (1969). Whether the wood elephant of Africa, a somewhat smaller animal, is a "good" species, remains uncertain. Starck (1995) showed a meticulous tree of the many elephant ancestors with their relatives. He stated that the ancestors have been traced back to the Eocene. Starck (1995) also provided an extensive anatomical comparison between the two species. Elephas separated from Loxodonta in the late Pliocene.

A closer kinship to Sirenia has also been suggested; perhaps the hyrax is related to elephants as well, as is often suggested. Based on their morphological studies, Gaeth et al. (1999) have suggested that African elephants may have had an aquatic ancestry. A comprehensive book gives many data on elephants to which the reader may refer (Eltringham, 1982).

African elephants may weigh around 7,500 kg, Asian males around 5,400 kg. Female Asiatic elephants may have tusks, and their ears are significantly smaller than those of the African species. Neonates weigh between 80 and 130 kg. Maximal life expectancy is around 70 years, primarily because of loss of the molars, which then disables nutrition of this herbivorous animal. Most wild elephants, however, succumb at a much younger age.

  African elephants in Africa.
  Asiatic elephant at San Diego Zoo.
2) General gestational data

Elephants usually have single offspring. Nevertheless, there are reliable records of twin births, with both neonates surviving. This has occurred in a specific family of African elephants but is somewhat unusual. Other references suggest that twins occur with a frequency of 1-2%. Flugger et al. (2001) presented physiological data and general medical experience on Asiatic elephant births. This work was published following a meeting of a group of veterinarians to discuss the past experience of elephant births, including sonography. The paper presented some detailed recommendations on elephant management. A comparison of reproductive successes and infant mortality in Asian elephants was published by Taylor & Poole (1998). Lang (1963) described in great detail the assisted birth of an Asian elephant at his zoo. The nutrition of elephants was reviewed by Hatt & Liesegang (2001). All of these papers all give adequate access to the older literature, and Eltringham (1982) provided additional literature.

The length of gestations that produced live births in Asian elephants was 600-692 days (others have suggested +/- 644 days). Our own findings are detailed at the concluding table of this chapter for both, African and Asian elephants. We found mostly 640-660 days as the length of normal gestation. Age of first pregnancy is generally around 9 years - other authors have suggested it to be later. When in captivity, no pregnancy has occurred before the age of 19 years. The development of pathologic uterine changes (leiomyomas, endometrial polyps) generally prevents gestation in later years.

The placenta is delivered in one half of births by 3 hours, and in 87% by 10 hours after the birth of the calf. Sonographic observations and progesterone determinations have enhanced pregnancy supervision. A comprehensive literature review of elephant publications is found in the (difficult-to-access) dissertation by Güßgen (1988).

3) Implantation

Nidation is central and superficial, with mesometrial orientation of the yolk sac and
the embryo being antimesometrial in location. The yolk sac disappears by the time that the fetal weight is 10 g (Mossman, 1987). Nevertheless, Cooper et al. (1964) described a persisting yolk sac to term. Mossman disputed that. One of our placentas from an African elephant at term also had what histologically appears to be yolk sac. It is shown following the next picture.

  Elephant placenta at about 11 months gestation. The zonary nature of the placenta is apparent, as is the marginal hematophagous area and the central band-like attachment to the endometrium which contains maternal vessels. (Kindly supplied by Dr. W.R. Allen, Cambridge, UK).
  Small yolk-sac-like structure in placental surface of African elephant placenta from case 2 in Table 1 at end of this chapter.
More data on earlier gestational observations will soon become known when the data from the presentation by Allen et al. (2001) is published. These investigators studied 43 pregnant uteri between 16 days and 21 months gestation. The ovaries contained 2-5 corpora lutea. Implantation began equatorially by replacing the endometrial epithelium. The end product was an endotheliochorial zonary placenta. Incubation of placental tissue with steroid precursors did not yield estrogen or progesterone, and extracts had no gonadotropic activity.
In follow-up, the full publication is now available (Allen et al., 2002) and must be read for its superior details of pregnant uteri throughout gestation. Of particular interest is also the finding in two post partum uteri of significant hemorrhages.

4) General characteristic of placenta

The elephant placenta is large, heavy and bulky. Its characteristics at term and after spontaneous delivery have only been described by Cooper et al. (1964). Prior to that description, some data from of culls of pregnant animals existed, especially the studies by Amoroso and Perry (1964). Because of superior fixation of their material and since it came from different stages of development, this paper must be studied by anyone interested in the elephant placenta. According to Ramsey (1975), elephants have an endotheliochorial placenta with some hemochorial regions. The placenta is said to be "villous", with the ("ectopic") subplacenta being labyrinthine (Ramsey, 1975). I have been unable to make these distinctions in our material. Assheton (1905) disputed the alleged similarity of the elephant placenta to the hyrax placenta and felt that it most resembled in its "plicate" trophoblastic structure the placenta of the sheep.

  The labyrinthine nature of the elephant placenta is evident. This is from the term placenta of "Thor" (No.1 in table at end of this chapter).
The elephant placenta has a more or less zonary shape, with the ring of villous tissue, which is divided into either three or four major portions, usually antimesometrially. This is seen in the next two photographs of a delivered placenta from an African elephant. There is great similarity between the two species' placenta. The minor differences that have been reported are questionable. Moreover, the fact that at least once the two species have hybridized to deliver a healthy young, suggests that their placentation must be extremely similar.

Aside from the massive nature of the elephant placenta, there is a pronounced girdle of brown and sometimes green degenerated material next to the insertion of the membranes. Observers usually call this the marginal hematoma. Because of this and the zonary nature of the elephant placenta, this has led to speculations of a relation to the carnivore placenta. While the green nature of this degenerated material suggests the presence of blood breakdown products, iron stains are only focally positive. That is insufficient to explain the green color, a fact that has also been true for dog placentas which I have studied. The alleged degenerating blood is said to derive from ruptured maternal capillaries of the endothelio-chorial labryrinth (Amoroso & Perry, 1964). Mossman (1987) stated that this is quite different from the marginal hematoma of carnivores whose blood is endometrial-derived. He believed the color is caused by biliverdin. These aspects, however, are not firmly settled. In the most recently observed organ, there was again a huge amount of peripheral green discoloration with most of the pigment being phagocytized in the large cylindrical trophoblast. It was again not stainable for iron (Prussian blue reaction) but granular and somewhat shiny. The nature of the pigment is still unknown but it does not have the quality of biliverdin, as suggested by Mossman, as it is too shiny for that usually dull-appearing pigment.

The free membranes of Asian elephants have a fine villous surface, which is not true of African elephant placentas. There is a very large allantoic sac with four parts whose connective tissue membrane has many blood vessels. In addition, many but not all placentas have pronounced "verrucae" (often called pustules) on the allantoic/chorionic surface.

Weights of several term and immature placentas observed by me are listed in the table near the end of this chapter; they vary between 8.3 and 22.2 kg. Flügger et al. (2001) provided similar weights of the placentas in eight pregnancies. The placentas of surviving young weighed between 9 and 21 kg, without cord and membranes. The free membranes, when being weighed separately once, were 2 kg (with an 11 kg placenta). The placenta of two live-born African elephants detailed by Soma & Kawakami (1991) weighed 13.5 and 10 kg respectively. Some additional data have become available from the records at Tierpark Berlin and were kindly supplied by Dr. J. Wisser. They are incorporated in the table. In addition, in August 2005, another placental observation was made available by Dr. Don Schlafer of Ithaca, NY. These data are also added to the table at the end. That calf weighed 140 kg and the placenta was 22 kg. The neonate did well despite numerous and very unusual projections on the allantoic surface. The usual pustules were replaced over the labyrinthine region by a large mass of verrucae, the longest of which was 10 cm long and weighed 24 g (!). These verrucae had a 1 cm diameter stalk and were covered by cuboidal epithelium. One third of them had branching structures.
  Delivered elephant placenta with detached membranes below. Note the marginal cord insertion and small white pustules in the center. Fetal surface.
  The same delivered placenta, maternal surface. The ring-shaped (zonary) nature is apparent. Note the green border around membrane insertion. Allantois top right.
  Cross-section through the thick placental tissue of a mature elephant placenta. Despite statements of the existence of major structural differences between superficial and maternal aspects, the term placenta is quite homogeneous.
  Edge of placenta to show the green bordering old hematoma.
  Insertion of umbilical cord, showing its three blood vessels.
  White, pearl-shaped "pustules" or "verrucae" on the allantoic surface; the amnion (below left) is free of them. These structure follow the fetal blood vessels.
  Old hematoma (green debris) peripheral to membrane insertion.
  Allantoic pustules on left; at right is the maternal aspect of the "free" membranes with very finely granular villous projections.
5) Details of barrier structure

The "barrier" is largely endotheliochorial in this complex labyrinthine zonary organ that encircles the uterine lumen. Contrary to some statements in the literature, I have not found hemochorial regions. Nor is there a true difference between maternal and fetal surface regions in the structure of the labyrinthine architecture. No decidua is present in the delivered organ. The trophoblast of the term delivered organ is single-layered and appears to be syncytial.
  Endotheliochorial relation of African elephant placenta (first case in table at the end of chapter, ("Thor"). E=endothelium; M=maternal blood; T=trophoblast; F=fetal connective tissue.
  Major maternal vessel and its relation to trophoblast and fetal connective tissue of African elephant placenta; from case 4 in Table 1 at end of this chapter. T=trophoblast; F=fetal connective tissue; E=endothelium.

6) Umbilical cord

The umbilical cord initially attaches mesometrially. It is long, between 4 and 6 cm in thickness, usually marginally attached in the term placenta, and it ruptures spontaneously with some bleeding. Blood clot may be attached to the severed end. The cord is probably not long enough, however, to enable the infant to be delivered while the placenta is still inside the uterus. Hence it must rupture or the placenta must emerge simultaneously. The umbilical cords I have observed have measured, at term, between 65 and 100 cm in the delivered organ, with little remaining on the newborns. Soma and Kawakami (1991), however, measured the cords of their two placentas as being 170 and 109 cm in length. Mossman (1987) indicated that the cords were short and he so drew this in his diagrams. I believe that this is in error. The umbilical cord generally has no twists and contains between three and four blood vessels that have a 1 cm thickness. The problem in deciding whether three or four vessels are present (and this differs in the literature as well), is due to the branching of vessels in the cord. Thus, it becomes important to state exactly where the sections were taken. This is further addressed in the paper by Davis & Benirschke (1991), which is appended with a table and photographs at the end of the chapter. The vessels, of course, are very stout (1 cm) and there are numerous vasa vasorum, much like in the dolphin cords. There is also a large allantoic duct.
In the last-delivered placenta of an African elephant, the 100 cm long umbilical cord was intact and had three large vessels and a large allantoic duct.

This most recently delivered placenta of an African elephant (2/2004) had three large vessels and the splitting of these vessels into six is apparent at arrow.

7) Uteroplacental circulation

In contrast to many other placentas, especially those of human gestations, the major blood vessels of the elephant placenta and their immediate derivatives contain elastic fibers. They are depicted in the publication of an Indian elephant studied by Hofman (1967). I know of no other relevant study than the anatomical descriptions of the generative tract by Perry (1964).

8) Extraplacental membranes

In most elephant placentas the most unusual feature is the presence of the large number of "verrucae" or "pustules" on the inner surface of the allantoic/chorionic membrane. Soma and Kawakami (1991) described additionally a polypoid structure with long pedicle that apparently arose from these pustules. Ramsey (1975) referred to the verrucae as "buttons" of the chorionic plate. They are composed of loose connective tissue with numerous blood vessels.
The allantoic sac is very large and has four "lobes". It contains fetal urine and is connected to the bladder by a large duct in the cord. Hippomanes have not been described. Several authors have commented on the presence of some villi on the "free" membranes in the Asiatic elephant and their absence in African animals (e.g. Mossman, 1987).

  Membranes of elephant placenta from "Kumi" (No 3 in Table 1 at end of this chapter). Amnion above, allantoic sac lining below.
9) Trophoblast external to barrier

There is no extraplacental trophoblast.

10) Endometrium

According to Mossman (1987), "atypical decidua" occurs in Proboscidea but there is no real "decidualization".

11) Various features

Perhaps the most unusual structural feature of elephants aside from tusk and proboscis is the absence of their pleural cavity. Fetuses have a perfectly normal mammalian development but, in the last month of gestation, the lung fuses with the chest wall and mediastinum (Beyer et al.1990). Thereafter, the space is composed of white, elastic connective tissue. Elephants also have no bronchial cartilage and possess no dental nerves. The proboscis is unique and essential for feeding and drinking, as well as for normal social interactions.

12) Endocrinology

The elephants have a bicornuate uterus with the ovaries being located unusually high in the abdomen (Starck, 1995). There is an ovarian bursa. Accessory corpora lutea develop during gestation. Aside from the paper by Perry (1964), Balke et al. (1988) presented anatomical descriptions of the female reproductive tract, and made reference to other publications. Their data came from 30 elephant culls in Africa and provided measurements, as well as weights.

Schwarzenberger et al. (2001) have presented endocrine data of musth and spermatogenesis in semi-wild Asiatic working elephants. Manual rectal stimulation enabled sperm collection and its evaluation. The review by Flugger et al. (2001) provides some guidance as to progesterone excretion during gestation. It can be ascertained from urine and blood, and appropriate values are given by Niemuller et al. (1993). Other references can be found in the publication by Balke et al. (1988), including semen characteristics. Much recent effort has been spent on sonography and artificial insemination, which is now often successful.

Numerous endocrine studies have been published, beginning with Short's (1966) observation of the ovaries and reproductive tracts of a female elephant. Plotka et al. (1975) determined that progesterone levels rose to 480 pg/ml in pregnancy but estrogen levels were not useful for monitoring. Ramsay et al. (1981), however, felt that urinary estrogen levels were useful with which for following cycles; these occur in approximately 22-day periods. The most recent study comes from Allen et al. (2001) and has been alluded to in section three; it is in abstract form so far only. Hess et al. (1983) also concluded that cycles could be followed by steroid measurements and asserted that progesterone levels fall dramatically prior to parturition. Their findings of a much longer cycle period (16 weeks) were validated by blood steroid level measurements of Brannian et al. (1988). Still other lengths were obtained in the study by Olsen et al. (1994), which suggested that steroid analyses may be useful for management. Pregnancy steroid and prolactin studies were reported by Brown & Lehnhardt (1995). As would be expected from other endocrine studies, Soma and Kawakami (1991) were unable to stain placental sections with antibodies to hCG. Many other studies are referred to that must be studied if further insight is desired. Estrus is 3-4 days.

Graham et al. (2002) developed an enzyme-immunoassay for the measurement of luteinizing hormone in elephants. This is an important advance for the captive breeding and assisted reproductive regimes now practiced in zoo elephants.

Musth in elephant bulls was studied by Ganswindt et al. (2002) by measurements of testosterone and epiandrosterone from urine and feces. There was good correlation of elevated androgen levels with musth and the authors suggested that this correlation may be useful in the management of bull elephants.

It has recently been shown that the common lack of cycling in female elephants is probably the result from hyperprolactinemia (Meyer et al., 2004). When cabergoline was given orally, the prolactin levels fell immediately in one test animal and cycling resumed. The proximate cause for hyperprolactinemia remains unknown.

13) Genetics

Both species have the same chromosome number (2n=56), but minor structural differences have been observed in their karyotypes (Houck et al., 2001). Only once has a male hybrid between the two species been recorded, at Chester Zoo. The mother was an Asian elephant, 22 years old. The newborn died in ten days from enterocolitis, weighing 166 lbs. (Howard, 1979). Nevertheless, the ability to hybridize indicates a close genetic relation between the species and very similar reproductive physiologic behavior.

Hartl et al. (1995) showed relatively small genetic diversity by studying proteins and enzyme coding loci in captive Asian elephants, and they also showed minor cytogenetic diversity (C-bands) in eleven individuals studied. Similarly, Fernando et al. (2000) showed only minor genetic diversity with mtDNA studies among 118 free-ranging Asian elephants, and there was little diversity when this was compared with the African species. They suggested this to mean a relatively "slow molecular clock".

Karyotypes of male and female African elephants; G-bands above, C-bands below.


G-banded karyotype of Asian elephants (Houck et al., 2001).

14) Immunology

No studies are known to us.

15) Pathological features

A major problem in captive elephants, especially the Asiatic elephants, has been the infection with an elephant-specific herpes virus from which many animals have succumbed (Richman et al., 1999; Burkhardt et al., 2001; Schaftenaar et al., 2001). The African elephant is considered to be the natural host of this virus and the virus causes only minor disease manifestations in that species. An apparently initially healthy 145 kg neonatal Asiatic bull elephant in Berlin suddenly died from this virus infection at age 268 days, with typical lesions exhibited post mortem. Viral antigen had been demonstrated by PCR in the placenta and in many organs at death. The placenta had shown only some umbilical cord inflammation (Ochs et al., 2001). Genital "papillomatosis" is also reported to be a feature of elephants in captivity.

Tuberculosis presents a significant threat to elephants in captivity (Wohlsein et al., 2001). Both M. hominis and M. bovis have been implicated. Many captive and also some wild elephants suffer tooth disorders, some of which may be life threatening (Fagan et al., 1999a;Wohlsein et al., 2001). Older elephant cows are prone to develop endometrial polyps (hyperplasia) and leiomyomas that limit reproduction. Starck (1995) enumerates the large number of parasites that have been reported in elephants.

16) Physiological data

Aside from their unusual dentition, the trunk and their specialized digestive tract, elephants have normally an obliterated pleural space. Embryos fuse their chest cavity in the last month of gestation (Beyer et al., 1990). The reason for this obliteration has been controversial. Moreover, elephants possess no nerves in their dental pulp (Fagan et al., 1999b). More basic physiologic aspects are discussed by Eltringham (1982).

17) Other resources

Cell strains of animals from both species are available from the cell bank at CRES at the San Diego Zoo by contacting Dr. Oliver Ryder at oryder@ucsd.edu.

18) Other features of interest

Because of the difficulty in accessing the article by Davis & Benirschke (1991), it is reprinted at the very end. It provides more quantitative data. More knowledge of the point of rupture of the umbilical cord would be of interest in future, as well as fetal hormone determination.


Aguirre, E.: Evolutionary history of the elephant. Science 164:1366-1376, 1969.

Allen, W.R., Mathias, S., Skidmore, J., Ford, M. and van Aarde, R.J.: Placentation in the African elephant. Placenta 22 (7): abstract P77, page A28, 2001. Reprod. Suppl. 60:105-116, 2002. II.Morphological changes in the uterus and placenta throughout gestation. Placenta 24:598617, 2003.

Amoroso, E.C. and Perry, J.S.: The foetal membranes and placenta of the African elephant (Loxodonta africana) Philosoph. Trans. Roy. Soc. London, Series B. 248:1-34, 1964.

Assheton, R. and Stevens, T.G.: Notes on the structure and the development of the elephant's placenta. Quart. J. Microsc. Sci. 49:1-37, 1905.

Assheton, R.: The morphology of the ungulate placenta, particularly the development of that organ in sheep, and notes upon the placenta of the elephant and hyrax. Philos. Trans. Roy. Soc. Series B, 198:143-220, 1906.

Balke, J.M.E., Boever, W.J., Ellersieck, M.R., Seal, U.S. and Smith, D.A.: Anatomy of the reproductive tract of the female African elephant (Loxodonta africana) with reference to development of techniques for artificial breeding. J. Reprod. Fertil. 84:485-492, 1988.

Beyer, C., Benirschke, K., Wissdorf, H. and Schoon, H.A.: Beitrag zur Problematik der Pleurahöhlenobliteration beim Elefantenfetus (Elephas maximus/Loxodonta africana). Sympos. Erkrank. der Zoo- und Wildtiere. Eskiltuna, Sweden. Akademie Verlag, Berlin. pp. 379-385, 1990.

Brannian, J.D., Griffin, F., Papkoff, H. and Terranova, P.F.: Short and long phases of progesterone secretion during the oestrous cycle of the African elephant (Loxodonta africana). J. Reprod. Fertil. 84:357-365, 1988.

Brown, J.L. and Lehnhardt, J.: Serum and urinary hormones during pregnancy and the peri- and postpartum period in an Asian elephant (Elephas maximus) Zoo Biol. 14:555-564, 1995.

Burkhardt, S., Ehlers, B., Goltz, M. Ochs, A., Wittstatt, U. and Hentschke, J.: Genetic and ultrastructural investigations for detection and classification of the elephant herpesvirus and prophylactic goals. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 157-160, 2001.

Cell strains of elephants available from CRES: http://www.sandiegozoo.org/conservation/cres_home.html. Please direct your inquiries to Dr. Oliver Ryder (oryder@ucsd.edu).

Cooper, R.A., Connell, R.S. and Wellings, S.R.: Placenta of the Indian elephant, Elephas indicus. Science 146:410-412, 1964.

Davis, S. and Benirschke, K.: Observations on the placenta of elephants. Sympos. Erkrankungen der Zoo- und Wildtiere, Liberic, CSR, Akademie Verlag, Berlin, pp.39-45, 1991.

Eltringham, S.K.: Elephants. Blandford Mammal Series. Blandford Press, Poole, Dorset, England, 1982.

Fagan, D.A., Oosterhuis, J.E. and Roocroft, A.: Significant dental disease in elephants. Verh. Ber. Erkr. Zootiere 39:15-133, 1999a.

Fagan, D.A., Benirschke, K., Simon, J.H.S. and Roocroft, A.: Elephant dental pulp tissue: Where are the nerves? J. Vet. Dent. 16:169-172, 1999b.

Fernando, P., Pfrender, M.E., Encalada, S.E. and Lande, R.: Mitochondrial DNA variation, phylogeography and population structure of the Asian elephant. Heredity 84:362-372, 2000.

Flugger, M., Goritz, F., Hermes, R., Isenbugel, E., Klarenbeek, W., Schaftenaar, W., Schaller, K. and Strauss, G.: Evaluation of physiological data and veterinary medical experience in 31 Asian elephant (Elephas maximus) births in six zoos. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 123-133, 2001.

Gaeth, A.P., Short, R.V. and Renfree, M.B.: The developing renal, reproductive, and respiratory systems of the African elephant suggest an aquatic ancestry. Proc. Natl. Acad. Sci. USA 96:5555-5558, 1999.

Ganswindt, A, Heinstermann, M. Borragan, S. AND Hodges, J.K.: Assessment of testicular endocrine function in captive African elephants by measurement of urinary and fecal androgens. Zoo Biol. 21:27-36, 2002.

Graham, L.H., Bolling, J., Miller, G., Pratt-Hawkes, N. and Joseph, S.: Enzyme-immunoassay for the measurement of luteinizing hormone in the serum of African elephants (Loxodonta africana). Zoo Biol. 21:403-408, 2002.

Güßgen, B.: Vergleichende Zusammenstellung der Literaturbefunde über die Anatomie des Indischen und Afrikanischen Elefanten als Grundlage für tierärztliches Handeln. Inaugural Dissertation, Veterinary School, Hannover, Germany, 1988.

Hartl, G.B., Kurt, F., Hemmer, W. and Nadlinger, K.: Electrophoretic and chromosomal variation in captive Asian elephants (Elephas maximus). Zoo Biol. 14:87-95, 1995.

Hatt, J.M. and Liesegang, A.: Nutrition of Asian elephants (Elephas maximus) in captivity - an overview and practical experience. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 117-122, 2001.

Hess, D.L., Schmidt, A.M. and Schmidt, M.J.: Reproductive cycle of the Asian elephant (Elephas maximus) in captivity. Biol. Reprod. 28:767-773, 1983.

Hofman, W.: Vergleichend-morphologische Untersuchungen über das Vorkommen elastischer Fasern in den Zottengefäßen von Geburtsplacenten. Acta anat. 66:67-77, 1967.

Houck, M., Kumamoto, A.T., Gallagher, D.S. and K. Benirschke: Comparative cytogenetics of the African elephant (Loxodonta africana) and Asiatic elephant (Elephas maximus). Cytogenet. Cell Genet. 93:249-252, 2001.

Howard, A.L.: "Motty" - birth of an African/Asian elephant at Chester Zoo. Elephant 1:36-41, 1979.

Lang, E.M.: Geburtshilfe bei einem indischen Elefanten. Acta Tropica 20:97-114, 1963.

Ludwig, K.S. and Baur, R.: Zur Struktur der Geburtsplacenta und der unreifen Placenta des afrikanischen Elefanten (Loxodonta africana) Acta Anat. 68:612(abstr.) 1967.

Meyer, J., Ball, R. and Brown, J.L.: Why are so many elephants in captivity not cycling? Evidence of an association between ovarian inactivity and hyperprolactinemia. JEMA (J. of the Elephant Managers Association) 15(1):9-13, 2004.

Mossman, H.W.: Vertebrate Fetal Membranes. MacMillan, Houndmills, 1987.

Niemuller, C.A., Shaw, H.J. and Hodges, J.K.: Non-invasive monitoring of ovarian function in Asian elephants (Elephas maximus) by measurement of urinary 5ß-pregnantriol. J. Reprod. Fertil. 99:617-625, 1993.

Ochs, A., Hildebrandt, T.B., Hentschke, J. and Lange, A.: Birth and hand rearing of an Asian elephant (Elephas maximus) at Berlin zoo - veterinary experiences. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 147-156, 2001.

Olsen, J.H., Chen, C.L., Boules, M.M., Morris, L.S. and Coville, B.R.: Determination of reproductive cyclicity and pregnancy in Asian elephants (Elephas maximus) by rapid radioimmunoassay of serum progesterone. J. Zoo Wildl. Med. 25:349-354, 1994.

Perry, J.S.: The structure and development of the reproductive organs of the female African elephant. Philosoph. Trans. Roy. Soc. London, Series B. 248:35-51, 1964.

Plotka, E.D., Seal, U.S., Schobert, E.E. and Schmoller, G.C.: Serum progesterone and estrogens in elephants. Endocrinol. 97:485-487, 1975.

Pucher, H.E., Stremme, C. and Schwarzenberger, F.: Penile paralysis in a semiwild ranging Asian elephant bull (Elephas maximus) in Vietnam - case report. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 177-181, 2001.

Ramsay, E.C., Lasley, B.L. and Stabenfeldt, G.H.: Monitoring the estrous cycle of the Asian elephant (Elephas maximus), using urinary estrogens. Amer. J. Vet. Res. 42:256-260, 1981.

Ramsey, E.M.: The Placenta of Laboratory Animals and Man. Holt, Rinehart and Winston, New York, 1975.

Richman, L.K., Montali, R.J., Garber, R.L., Kennedy, M.A., Lehnhardt, J., Hildebrandt, T., Schmitt, D., Hardy, D., Alcendor, D.J. and Hayward, G.S.: Novel endotheliotropic herpesviruses fatal for Asian and African elephants. Science 283:1171-1176, 1999.

Schaftenaar, W., Mensink, J.M.C.H., de Boer, A.M., Hildebrandt, T.B. and Fickel, J.: Successful treatment of a subadult Asian elephant bull (Elephas maximus) infected with the endotheliotropic elephant herpes virus. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 141-146, 2001.

Schwarzenberger, F., Pucher, H.E., Stremme, C., Holzmann, A., Tu, N.C. and Ly, C.T.: Endocrinological studies in semi-wild ranging male Asian elephants (Elephas maximus) in Vietnam: Correlation to musth and spermatological parameters. In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 135-139, 2001.

Short, R.V.: Oestrous behaviour, ovulation and the formation of the corpus luteum in the African elephant, Loxodonta Africana. East Afr. Wildl. J. 4:56-68, 1966.

Soma, H. and Kawakami, S.: Parturition and placentation in African elephant (Loxodonta africana). Verh. Ber. Erkrank. der Zoo- und Wildtiere. Akademie Verlag, Berlin, pp. 47-51, 1991.

Starck, D.: Lehrbuch der Speziellen Zoologie, Vol. II, part 5/2. Gustav Fischer Verlag, Jena, 1995.

Taylor, V.J. and Poole, T.B.: Captive breeding and infant mortality in Asian elephants: A comparison between twenty Western zoos and three Eastern elephant center. Zoo Biol. 17:311-332, 1998.

Wohlsein, P., Bartmann, C.P., Kirpal, G., Hildebrandt, T. and Peters, M.: Tuberculosis, molar malocclusion and urogenital lesions in a captive Asian elephant (Elephas maximus). In, Proceedings of the Institute for Zoo and Wildlife Research, H. Hofer, ed., #4, pp. 161-171, 2001.


Because of the difficulty in accessing this article, it is here reprinted in full.

University of California and the San Diego Zoo, San Diego, California
By S. Davis and K. Benirschke

(Citation: Davis S. and Benirschke, K. Observations on the placenta of elephants. Symposium on Erkrankungen der Zoo- und Wildtiere. Liberic, CSR. Akademie Verlag, Berlin, pp. 39-45,1991).


The placenta of elephants has not often been studied, and delivered specimens from mature Asian elephants have only been described by COOPER at al. (1964). Especially lacking are studies that compare the placentas of the two species with one another. The most detailed report of implanted specimens from African elephants at various gestational ages comes from the pen of AMOROSO and PERRY (1964). These authors had incomparably fixed uteri and made precise descriptions of the various structures they identified in the elephant placenta. They preferred to call the type of placentation in this species as being "vasochorial", rather than the conventional "endotheliochorial" type. In his comprehensive book on the placentas of all mammals, MOSSMAN (1987) considered the elephant placenta as belonging to the "hyracoid type", as being incompletely zonary, labyrinthine and endotheliochorial, with large allantoic sac, small marginal hematoma, and with accessory villous areas over an otherwise smooth allantois. His drawings suggest that the allantoic sac extends into the empty uterine horn and that the free amnio-allantoic membrane contains fetal blood vessels. MOSSMAN (1987) also pointed out that the similarity of the elephant placenta to that of carnivores (initially suggested by the zonary shape and marginal apparent hematomas) is incorrect. The placenta of elephants, however, is believed to have great similarity to that of the hyrax and manatee in that all have fourlobed allantoic sacs, amnionic and allantoic verrucae (often called pustules), and short umbilical cords in which veins and arteries were alluded to as being paired. AMOROSO and PERRY (1964) found no evidence to suggest that the zonary shape of the placenta results from atrophy of portions of a formerly diffuse placenta. They could not affirm the presence of minor villous projections on the free membranes, findings that are in agreement with ours.

At San Diego Zoo's Wild Animal Park groups of female African and Asian elephants have been kept and, in recent years, males of both species have been acquired. Since then, five African and three Asian elephants have been born, in addition to an abortus of an Asian elephant. The placentas were obtained from all within few hours after delivery. They were weighed, measured and gross observations were recorded. Many photographs were taken. Histologic sections of defined areas of placenta, membranes and umbilical cord were prepared. These findings represent the substance of this report.


The principal macroscopic findings are summarized in Table I and a brief description is given here of the individual cases.

Case 1. The umbilical cord was 59 am long with three vessels. It divided into three portions at this point, measuring 4,B, 49 and 66 cm respectively until these entered the placenta at the margin (Fig. 1). Thus the total cord length was 125 cm (The length of all the umbilical cords was calculated as the longest distance from cord rupture site to the entry of the cord into the placental bed). The placenta had two large lobes, separated by 68 cm of membranes. The lobes measured 70 x 28 x 6 cm and 62 x 22 x 1O cm. A marginal hematoma and green staining of the membranes was noted. Many allantoic pustules were present. The 95.5 kg infant was gored by an adult female at age 4 days and died one week later.

Case 2. The total length of umbilical cord was 112 cm. The placenta had two lobes weighing 5,500 and 3,200 g respectively. The margins of the placenta had the usual hematoma, and allantoic pustules were present. Delivery of the calf was not observed, thus the dam had eaten approximately half of the placenta prior to examination.

Case 3. The placenta of this 13 year old African elephant was ring-shaped, composed of three separate lobes and weighed 22.2.kg. The lobes measured 44 x 26 cm and 44 x 28 x 4 cm. There was a green marginal hematoma, measuring up to 0.3 cm in thickness. The allantois was covered with pustules. The cord measured 65 cm in length. It had three vessels, but 43 cm from the torn end, one vessel branched. The cord also contained a huge allantoic duct and measured 6 cm in diameter. Pustules and hematoma of the spontaneously ruptured umbilical cord are shown in Figure 2. The mother had been immobilized several times during pregnancy for surgical repair of a broken leg.

Case 4. The umbilical cord inserted in almost marginal fashion and measured 65 cm. The vascular division began approximately 30 cm from its fetal end. It contained a large allantoic duct and three vessels. Abundant pustules were present on the membranes (Fig. 3). The placenta weighed 13.45 kg. There were three lobes measuring 34 x 35 x 5 cm, 31 x 41 x 4 cm, and 47 x 22 x 4.5 cm, weighing 6,000, 4,250 and 3,200 g respectively. A marginal hematoma was present, measuring up to 1 cm in thickness.

Case 5. The umbilical cord of this 11 year old African elephant's baby was 100 cm long, possessed three blood vessels and large allantoic duct. Its fetal end had much hemorrhage. The placenta was ring-shaped but had 3 ovoid lobes and weighed 17.6 kg. Relatively few pustules were observed in this specimen; there was a marginal hematoma at the edge of all three lobes.

Case 6. The umbilical cord measures 50 cm in length and the four fused lobes of placenta, which had a generally annular shape, weighed 8.3 kg. Its thickness was 5 cm. There were no pustules on its shiny surface. A marginal hematoma was present. The membranes which were weighed separately, were 1,350 g. The cord weighed 750 g and, approximately 9 cm from its fetal end, the vein divided and 13 cm further, another branching took place. Eventually, four artery/vein pairs entered the placental surface.

Case 7. The placenta of this stillborn calf weighed 1,100 g with its attached 23 cm long cord. The latter had three twists and contained 2 arteries and one vein, as well an allantoic duct that funneled into the allantoic cavity. The allantoic sac was covered with brown pustules in its center, varying in size up to 2 cm and 0.2 cm in thickness. They disappeared towards the edge of the placenta. There was no marginal hematoma. Histologically, this placenta was markedly different from the others. Its blood vessels contained calcified former thrombi and these extended deeply into the placental tissue. The maternal vessels had much more recently developed thrombi and the interdigitation of maternal and fetal tissues was very evident. Moreover, the uterine epithelium had not atrophied to produce a vaso-chorial placenta, and the trophoblastic coverings of the fetal placental trabeculae (lamellae) was degenerated.

Case 8. This placenta was delivered at Cesarean section from a 19 year old nulliparous Asian elephant. The dam had a 744 day gestation based on observed copulation and subsequent rise in progesterone. A progesterone level 76 days prior to delivery had fallen to baseline. It was determined that the calf had died and a decision was made to perform a Cesarean section to attempt to salvage the dam. The placenta weighed 11.16 kg, and the membranes 2.12 kg. The umbilical cord was 82 cm long and 4 cm thick. The annular placenta measured 14 x 34 x 3 cm.

Case 9. This stillborn' a placenta weighed 14.5 kg. The dam had ruptured the membranes 5 days prior to delivery of a large stillborn calf.


Table 1: Macroscopic findings of elephant placentas
Weight of
Weight of
Length Remarks
of Cord
L. africana          
No.1 (4356) %, live
639 d 95.5 kg 21.8 kg 125 cm Died,
No.2 (4432) %, live
666 d ? 8.7 kg 112 cm,
Green border
No.3 (4511) %, live
?? 128.1 kg 22.2 kg 65 cm Died,
28 days
No.4 (4529) %, live
644 d ? 17.6 kg 65 cm,
Three lobes
No.5 (4522) &, live
662 d ? 17.6 kg 100 cm,
Three lobes
Additional male: February 2004. (male surviving neonate)       13.7 kg 104 x 6 cm. 3 vessels
No.7 (S-8215) “Mabu” live   95 kg 9.25 kg 100 x 4 cm. 3 vessels
E. maximus          
No.6 P90-582 %, live
664 d 120.5 kg 8.3 kg 50 cm
Four lobes
No.7 P89-571 &,abortus 10m 3.05 kg 1.1 kg 23 cm
3 twists cord
No.8 &, stillborn
of "Jean"
744 d 131.3 kg 11.2 kg 82 cm Cesar.
No.9 &, stillborn 666 d 151.3 kg 14.5 kg ? Rupt.
sac-5 d
No. 10 premature
Tierpark Berlin
  +/- 60 kg 8.75 kg
86 cm Allantois
90x45 cm Size: 50x26 cm, 3-3.5 cm thickness
No. 11 2001
Tierpark Berlin
  110 kg 25.5 kg 122 cm Allantois: 104x70 cm. Size: 66x30 cm, 3.8-5.2 cm thickness
No. 12 Munich Zoo     11.1 kg Allantois: 80x25 cm; Size: 30x3.5 cm
No. 13 Syracuse Zoo     140 kg 22 kg 200x115 cm, girdle 40x80 cm, cord 90 cm long, 6% of weight, unusually long allantoic verrucae

The gestational ages given are based on observed copulation and/or progesterone elevation.
+ Placenta half-eaten before retrieval.
++ Progesterone decline to baseline 76 days before delivery.
The data of ## 10-12 were kindly supplied by Dr. J. Wisser and Hildebrandt.


It can be seen that the weights of full-term African elephants placentas is generally greater than that of their Asian counterparts. The weights in our Asian specimens compare favorably with those of the two placentas reported by COOPER at al. (1964), who found weights of 11.3 and 11.8 kg at term. The lengths of umbilical cords given by these authors (110 and 114 cm), however, is greater than those observed by us. Whichever the case may be, the total length of the umbilical cord is shorter than that of the vaginal canal. Therefore, rupture of the cord would have to occur during delivery of the calf, as the placenta is not usually delivered simultaneously with the calf. The resultant hematoma at the site of rupture was often observed in our specimens. It has recently been suggested that the length of the umbilical cord is a function of fetal mobility (review BENIRSCHKE and KAUFMANN, 1990). This is also correlated with the degree of twisting in the cord, absent in elephants and in most other species in which the fetus is longitudinally oriented. It is thus impossible for the cord to become twisted, except in very early stages of development and, for this reason, we believe that the twisting of the umbilical cord found in our specimen No. 7 (Table 1) may have occurred early in gestation and was the probable reason for that fetus' death.

The umbilical cord, which always contained an allantoic duct, often has a nearly central insertion on the placenta, but this is not always the case (Fig. 2). It is of impressive thickness, usually about 5 cm in diameter and, at its site of spontaneous rupture, a hematoma may develop. As did COOPER et al. (1964), we have observed that the umbilical cord of elephants possesses three vessels (2AIIV) which, after branching become pairs of arteries and veins as they approach the placental surface, while AMOROSO and PERRY (1964) identified 2 veins and 2 arteries in African elephants, a difference that is not reconciled since they did not specify the point where sections were taken. Perhaps they took these closer to the placental surface than other observers. The diameter of the umbilical vessels is individually, approximately 1 cm, and all vessels possess vasa vasorum, in contrast to cord vessels of many other species.

The surface of the richly vascularized chorio-allantoic membrane that is apposed to the uterine wall contained numerous capillaries, while the amnion was devoid of same. On the other hand, the "inner" membranes, those which ASSHETON and STEVENS (1905) described as not "attaining the inner surface of the placenta" had no blood vessels in our observations, which is contrary to the drawing of MOSSMAN (1987; his Plates 25, 26). The epithelium of the amnionic cavity was degenerated in all of our cases, while that of the allantoic sac was flat or transitional in nature. Amnion and allantois possessed the numerous polypoid projections that have been referred to as verrucas or "pustules" (AMOROSO and PERRY 1964). As depicted by them, the pustules are constructed of proliferations of connective tissue and small blood vessels and vary considerably from polyp to polyp. In some sections the pustules appeared to commence by a proliferation of small vessels beneath papillary projections from the allantoic sac. Their function is unknown; they follow strictly the course of larger fetal surface vessels, and they do not bear any similarity to the keratinized patches of cord and amnion found in many ungulates and cetacea. At the periphery of the placenta, these pustules become smaller and then disappear. Remnants of yolk sac were not identified.

We found only two typical, completely annular placentas, but observed that most placentas were made up of several lobes. AMOROSO and PERRY (1964), who examined eight placentations of African elephants in great detail, also suggested that the placenta is "more usually interrupted at one or more points on the circumference". The entire expanse of a term placenta is quite variable and this is summarized in above observations. It is a thick (5 - 6 cm), flashy organ and not easily histologically sectioned completely because of its size. The superficial vessels have an impressive thickness.

At the edge of the placenta, a 6 - 8 cm wide brown/green discoloration was present in most placentas. While it resembled old blood clot, this may only be a superficial macroscopic observation. This "hematoma" bears some similarity to the marginal hematoma of carnivores, but it differs in many ways. The marginal "hematoma" of brown-green nature has been remarked upon by almost all observers and is referred too by ASSHETON and STEVENS (1905) as being made up of dark brown and yellow pigment granules of varying sizes. When using special stains, these authors already found the pigment to be free of iron and made other histological studies that failed to reveal the identity of the pigment completely. They considered it to be an excretory product. ASSHETON (1906) also found by special stains no iron to be deposited in this location as was the case in our preparations. Rather, the iron was located in the adjacent fetal connective tissue and the iron-staining took place mostly in the perivascular spaces. This observation of absent hemosiderin granules conforms to our special stains that also sought to identify the pigment. The brown granules also did not react with stains designed to disclose bile pigments. It was, however, markedly PAS positive and accumulated in the lumens of spaces that are lined with tell cytotrophoblast. The granules were smallest and least numerous in this columnar trophoblastic epithelium. COOPER et al., (1964) suggested that "in the marginal pigmented zone…. are villi .... apparently defunctionalized areas (that) probably arise by extravasation of maternal blood with degeneration of maternal structures". But we believe that this is not consistent with the finding of absent hemosiderin and hematoidin pigments in this granular pigment. AMOROSO and PERRY (1964) considered the pigment to derive from blood issuing from capillaries of the placental labyrinth, rather than from maternal vessels, but their identification of it as of blood-derived material is likewise incomplete. They liken it to histotrophe and infer a nutritive quality to this material.

The histologic features of the elephant placenta have been described in great detail by ASSHETON and STEVENS (1905), who considered the elephant's placenta as deciduous (with little maternal tissue being shed) and they also reviewed the sporadic earlier reports in some detail. ASSHETON (1905) described the placenta as being "closely comparable to that of the sheep. The placenta is essentially plicate...". The specimens examined by these authors and described in greater detail by ASSHETON (1906) subsequently, however, were probably not of optimal fixation and their observations, therefore, lack decisive detail and have been criticized subsequently, especially the maternal vascular supply.

Much better preparations were made by AMOROSO and PERRY (1964) who depicted and described the histology in excellent detail. COOPER et al. (1964) considered the elephant placenta to represent a labyrinthine endotheliochorial organ and they differ with previous descriptions that suggested a hemochorial placentation, a finding that is in accord with more recent observations and ours as well. The fetal vessels enter finely divided lamellae which are covered with a thin layer of trophoblast and appose the denuded maternal capillaries. It is thus endothelio-(vaso)-chorial. But the architecture differs at the different levels of the placental depth (zones I and II of AMOROSO and PERRY). Thus, towards the maternal surface, there is a sharp separation from tissue that large cellular trophoblast.. Little degenerating endometrium is shed with the delivered placenta. We observed calcifications in all of our term elephant placentas. Since histological details are well described by AMOROSO and PERRY (1964) who had clearly the best-preserved tissues to examine, we will not repeat these findings here. Their observations were also supported by subsequent studies of LUDWIG and BAUR (1967).


The cooperation of Mr. Alan Roocroft, San Diego Wild Animal Park, is gratefully acknowledged.


Observations on the Placenta of Elephants

We here describe the macroscopic features of 5 African and 4 Asian elephants' placentas and accessory structure. While the placentas of African elephants have a somewhat greater weight, the placentas of the two species show great similarity and differ only in minor details. Histological observations confirm those previously published, except that the umbilical cord possesses 3 rather than 4 vessels as suggested by some previous students of the elephant placenta.


AMOROSO, E.C. and J.B. PERRY (1964).The foetal membranes and placenta of the African elephant (Loxodonta africana). Philos. Trans. Roy. Soc. London, Series B, 248, 1-34.

ASSHETON, E. (1905). The morphology of the ungulate placenta, particularly the development of the organ in the sheep, and notes upon the placenta of the elephant and hyrax. Roy. Soc. London Proc, 76, 393-394.

ASSHETON, E. (1906). The morphology of the ungulate placenta, particularly the development of the organ in the sheep, and notes upon the placenta of the elephant and hyrax. Philosoph. Trans. B., 198,143-220.

ASSHETON, R. and T.G. STEVENS (1905). Notes on the structure and the development of the elephant's placenta. Quart. J. Microscop. Sci. 49, 1-37.

BENIRSCHKE, K. and P. KAUFMANN (1990). The Pathology of the Human Placenta. New York Springer.

COOPER. I., CONNELL, R.S., and S.R. WELLINGS (1964). Placenta of the Indian elephant, Elephas indicus. Science 146, 410-412.

LUDWIG, K.S. and R. BAUR (1967). Zur Struktur der Geburtsplacenta und der unreifen Placenta des afrikanischen Elefanten (Loxodonta africana). Acta Anat. 68, 612 (abstr.)

MOSSMAN, H.W. (l987) Vertebrate Placental Membranes. London The MacMillan Press Ltd.

  Figure 1: Placenta of case 1. African elephant. Note the marginal insertion of the umbilical cord, the hematoma near its rupture, and the vascular division with paired vessels entering the placental tissue.
  Figure 2: Case 3. African elephant placenta. The paired umbilical vessels coursing from the lateral cord insertion are apparent. Note the many allantoic pustules (white) and at left the reflected amnionic, allantoic sacs.
  Figure 3. Case 4. Delivered placenta of African elephant. The umbilical cord is, again, inserted at the margin of the placenta with disruption and hematoma at the fetal end. The white pustules generally follow the major fetal vessels.
back to top

© 2007.
welcome | home | index | intro | placentation | glossary | author | contact