Seasonal variation in breeding in I. macrourus was observed both in captivity and in the wild in Brisbane. Although animals with young were observed throughout the year, the main period of breeding occurred from late July (early spring) through summer, until March (autumn). This seasonal production of young was more noticeable in those animals in the outside enclosure, since all the 32 pregnancies recorded over the three years of study occurred during the summer period. A fertilisation rate of 58% (n = 38) was determined for these captive bandicoots, the rate being calculated as the percentage of matings that resulted in conception.
Unlike most other marsupials, the bandicoots have corpora lutea of pregnancy which persist into lactation. Morphologically, in I. macrourus, the luteal tissue is present until approximately day 42 of lactation. On examination of the plasma progesterone concentrations, these were found to increase from 3.9 to 12.6 ng/ml between 11 and 4 days before parturition, and then remained at this level for at least a further 7 days. The plasma progesterone levels gradually declined until basal concentrations were reached at approximately day 19 of lactation. It would appear that although the corpora lutea of pregnancy persist for a further 20-25 days of lactation, they are not at this stage involved in the production of progesterone. This has been further substantiated by ultrastructural examination of luteal tissue during pregnancy and lactation. Dense staining granules, thought to be implicated in progesterone secretion in other mammalian species, were observed in luteal tissue from day 5 of pregnancy until day 9 of lactation coinciding with active hormone secretion, and were absent when secretion had ceased on day 25 of lactation.
As stated previously, the plasma concentration of progesterone did not alter significantly from 4 days before parturition, during parturition and up to 3 days of lactation. On the other hand, the concentration of 13,14-dihydro-15-keto-prostaglandin F2α (PGFM) increased from basal levels of 288 pg/ml 4 days before birth, to maximal levels of 2534 pg/ml immediately before parturition, then decreased progressively. Thus, the bandicoot differs from the majority of mammals in giving birth when progesterone concentrations in the maternal plasma are elevated, whereas in most eutherian mammals so far studied, a drop in the plasma concentration of progesterone precedes birth. It is therefore suggested that PGF2α, and not progesterone initiates parturition in this species.
When the plasma concentration of progesterone in two bandicoots was monitored following the removal of the young early in lactation, it was estimated that ovulation occurred on day 21 and day 24 of lactation respectively. After surgical removal of the corpora lutea along with removal of the young, ovulation was estimated to have occurred on day 11 in one bandicoot, and between day 10 and 15 of lactation in another. Therefore it would appear that when actively secreting progesterone, the corpora lutea have an inhibitory effect on ovulation. If the corpora lutea were removed, and the young allowed to continue suckling, then ovulation was inhibited (n = 2) suggesting that suckling may inhibit follicular growth and ovulation throughout lactation.
In conclusion, it is known that bandicoots differ from other marsupials in that they possess chorio-allantoic placentae, have a very short gestation period of 12½ days and have corpora lutea of pregnancy which persist into lactation. This study provides further evidence of the bandicoot's uniqueness in that parturition takes place without a decrease in plasma progesterone, and the corpora lutea, which are present for approximately 42 days after birth, secrete progesterone during the first half of lactation only.