Back to Tenrec Resources and Information
The reproduction and rearing of tenrecoid insectivores in captivity
John Eisenberg & Nancy Muckenhirn
National Zoological Park, Smithsonian Institution, Washington, DC, USA
For the last two years the National Zoological Park, Washington, has maintained a rather extensive collection of tenrecs. As part of our research programme, we have supported both field investigations in Madagascar and laboratory research on the biology of tenrecs. A central problem in our studies has been the successful captive maintenance and breeding of these rare insectivores. Initial results on the general biology and captive maintenance may be found by referring to Eisenberg and Gould (1967) and Gould and Eisenberg (1966). In the present paper, we wish to present some of our current results concerning the captive propagation of the subfamily Tenrecinae including: Tenrec, Setifer, Echinops and Hemicentetes. The remaining subfamily, the Oryzorictinae, is at present under study and will be the subject of a later report.
In addition to diurnal periodicities in body temperature, all species of the Tenrecinae exhibit seasonal periods of torpor, generally during the austral winter (Gould and Eisenberg, 1966). Before their entry into the torpor period, individuals show an increased appetite and rapid fat deposition. In our laboratories all newly captured specimens continue to maintain during their first year at Washington their annual cycle with respect to a southern hemisphere pattern. After four years in captivity, our Echinops colony is finally synchronised to breed during the northern spring. Our other stocks are still in the process of gradual adaptation to the reversed annual cycle.
In order to promote the reduction in accumulated fat reserves and allow the specimens to complete their annual cycle in accordance with their periodic physiological rhythms, we now place all our specimens in rooms maintained at approximately 16°C for three to four months. During this time the animals are exposed to lowered ambient temperatures, they are given only fresh water and all food is withheld. Their weigths are checked at intervals until an individual falls to a level equal to that noted in the field during their normal active phase. At this time the animal is returned to normal food rations and moved to heated quarters (Eisenberg and Gould, 1967). It should be noted, however, that all species of Tenrecinae will show torpor, even at ambient temperatures of 20-24°C. At these temperatures spontaneous arousal is of frequent occurrence, but the arrangement has been satisfactory for Echinops.
In our experience E. telfairi, Hemicentetes nigriceps, and Tenrec ecaudatus exhibit a relatively prolonged period of torpor with their rectal temperatures remaining near the ambient. H. semispinosus and Setifer setosus are more likely to exhibit periods of arousal for several days at a time with a concomitant rise in rectal temperature coupled with active movement about their cages.
Although many of our tenrecs are allowed to hibernate as family groups, only species of the genus Hemicentetes are held as groups during the breeding season. Specimens of Echinops, Tenrec, and Setifer are maintained in isolation before pairing. Pairing in Tenrec has been accomplished in cages having a floor area of either 8 or 3 sq. m. The other genera have been paired in cages with 0.43, 0.25, and 3.0 sq. m floor space. During the pairing interval nest boxes are provided. Tenrec will copulate in the open but the other species invariably copulate in a nest box.
After emerging from torpor, the animals gain weight rapidly and exhibit mating behaviour in about two weeks. Females vary in their receptivity but after several days with a male, mating is generally accomplished. When first placed together, a male generally attempts contact and, if not immediately repulsed, he will attempt to mount (Gould and Eisenberg, 1966). The mount with intromission is prolonged, lasting up to 25 minutes, and may be followed by repeated mounting during the following days. We generally separate a pair following a series of successful series of mounts but if these animals are, as we believe, induced ovulators then the repetitive mount series over a 24-hour period may be essential for successful ovulation and implantation.
Separation of the male from the female after mating appears essential for the successful breeding of Setifer setosus and Tenrec ecaudatus. Females of these species appear very sensitive to any disturbances before parturition and abortion or resorption of embryos is not uncommon in our colony. Although H. semispinosus is quite tolerant and even forms extensive family groups in the same burrow system (Gould and Eisenberg, 1966), there is nevertheless a marked tendency for the female to withdraw into a separate nest box at partus and for this reason we now separate all pregnant females of this species. With the more solitary H. nigriceps, this requirement appears to be even more important.
The nest boxes for Tenrec have a central chamber measuring 45 by 29 by 20 cm, with a connecting tunnel 20 by 13 by 27.5 cm long. The other species use a smaller box of similar construction having a chamber 30 by 15 by 10 cm, with a 9 by 7 1/2 by 10 cm tunnel. Since Hemicentetes and Tenrec in nature tend to defaecate at the entrance to their tunnel, we have experienced some difficulty with these tunnel boxes. In captivity the specimens frequently defaecate and urinate in the wooden tunnel, complicating problems of sanitation, especially when young are in the box. We now recommend that the tunnel be eliminated for those females which develop the habit of not actually leaving the nest box to defaecate. The nest chamber itself could be reduced to a 25 by 20 cm floor area for breeding Tenrec, since the larger chamber permits a new-born litter to become scattered and reduces the chance of successful rearing.
New data concerning gestation and litter sizes are included in Tables 1 and 2. The long gestation of the Tenrecinae helps in part to account for their rapid maturation which was noted in a previous publication (Gould and Eisenberg, 1966). As can be noted in Table 2, our captive-bred litters of T. ecaudatus are smaller in number than those obtained in the wild, however, captive breedings of other species are comparable to the field data.
|Hemicentetes semispinosus||2||> 45 <= 55|
|H. nigriceps||--||(as for H. semispinosus)|
|Size of litter
|Size of litter
in the wild
(Mode = 7)
The suspectibility of Setifer and Tenrec females to disturbance resulted in poor rearing by the mothers. As a result, hand rearing techniques were attempted.
Four common tenrecs T. ecaudatus were born on 23 February 1967 to a female which had been unsuccessful in rearing a previous litter of two young. Since one young of this second litter did not survive the first day, two siblings were taken for hand rearing. The birth weights were 22.8 and 27.4 g and the total lengths 84 and 88 mm. The average birth weights and lengths for the previous litters of the species were 25.2 g and 85.2 mm, respectively. The two young appeared to be doing well on 2-hour feedings of equal parts of Carnation evaporated milk (Borden Co) and water, but died two days after birth as an apparent result of temperature stress resulting from an accidentally large fluctuation.
The last surviving young was found outside the nest box on the sixth day and weighed only 18.2 g. It was successfully hand-reared on a 2 : 5 : 1 mixture of Esbilac : water : Carnation evaporated milk. Lactinex (Hynson, Wostcott and Dunning, Inc., Baltimore, Maryland) was added as a dietary supplement for the natural bacteria removed in the process of milk sterilisation. The young was kept in an incubator at 26.7°C during its first month of life.
Feeds were given every two hours during the second week until the young regained the birth weight of its siblings and then at 3-hour intervals until it ate solid food at 3 1/2 weeks of age. Between 7 and 14 g of milk was generally accepted from an eye dropper per feeding, totalling 56 to 113 g consumed daily. The tenrec defaecated spontaneously. Unmetabolised fat was found in the faeces. The tenrec was weaned to a diet of chopped earthworms and horsemeat, including supplemental vitamins (ViDaylin-M of Abbot Laboratories, North Chicago, Illinois) and a 2 : 1 Purnival (US Vitamin & Pharmaceutical Corp., New York, N.Y.) : calcium carbonate mixture.
A Giant hedgehog tenrec S. setosus was also hand-reared on a similar diet. It was removed from the female on the fourth day after a 1 g weight loss from 21.7 g on the third day. Two modifications were introduced for rearing: since the animal was cool and inactive when incubated at 25.6 to 26.7°C, a higher incubator temperature of 29.4°C was employed; and secondly, different dietary supplements were used. Zymadrops (Upjohn Co), which has a higher vitamin D content, was substituted and dicalcium phosphate (DCP 340, Parke, Davis & Co, Detroit, Michigan) at 0.1 to 0.2 g/oz milk was added to the diet to curb rickets which developed by the fifth week. The diet was also restricted to reduce the obesity which hindered use of the hind limbs. It is suggested that the leanest possible meat be used in the diet. The animal regained considerable support and coordination of its hind legs by the 10th week with this treatment.
Maturation of the young
Detailed discussions of growth and development for H. semispinosus and E. telfairi are included in Gould and Eisenberg (1966). We include here rather detailed notes on the growth and maturation of Tenrec and Setifer. A summary of the salient features of tenrec maturation is included in Table 3.
|Earliest age of
|Tenrec ecaudatus||1||9-15||25-30||Following season = 6 mos.|
|Echinops telfairi||18||7-9||18-22||Following season = 6 mos.|
|Setifer setosus||2||9-13||15-20||Following season = 6 mos.|
The eyes opened between 9-15 days of age. The tenrec began a moult to the subadult pelage during the third week. In the striped pattern of the young, five rows of white spines ran longitudinally along the black dorsum. These spines, which were 1 mm long at birth, reached a maximum length of 5 mm in the third week, became bent and finally lost by the 10th week. Specialised, stout, stridulating spines occurred posteriorly. As the animal grew, the spines became increasingly spaced as indicated by the length of the area which increased from 15 to 75 mm by week 10. These spines produce an audible sound when rubbed together (Gould, 1966) and were gradually covered by the growth of a reddish brown coat which varied from coarse hair posteriorly to a shorter, more spinescent hair anteriorly. The behaviour pattern of drawing the neck muscles forward was present as a 'startle' reaction long before the nuchal crest spines, which are fanned in startled adults, grew. At birth the animal was 84 mm long and weighed about 25 g; at 26 days of age, it had reached a total length of 106 mm and weighed over 50 g. At the end of the second month, this animal weighed 440 g and had reached 265 mm in total length.
The eyes began to open on day 9 and were both completely open by day 14. The tenrec began to eat solid food on day 15. There was no distinct moult to a subadult pelage and the spines grew continuously from day one. The young was first measured and weighed at three days of age. Total length equalled 79 mm and the weight totalled 21.7 g. By two weeks, the young Setifer had reached 113 mm total length and weighed 29.7 g. By one month this animal measured 127 mm and at two month 140 mm total length with a weight of 101 g.
Discussion and summary
This paper presents a summary of our current efforts to breed five species of the family Tenrecidae. Certain problems have been encountered concerning the low fertility of T. ecaudatus and the breakdown of maternal care in both S. setosus and T. ecaudatus. Some of these problems may be related to the structure of the nest boxes and we are trying to use a newly designed box which will provide the female with maximum security. Another aspect of the problem may be related to general obesity and a metabolic decline during pregnancy when we feed our females to satiation. On the other hand reduced rations may induce early torpor in specimens with extensive fat deposition. In the coming year, we plan to institute more rigorous control over food intake during the early phases of pregnancy in an effort to reduce the variables which appear to affect adversely our success with this group.
Our captive breeding experiments indicate a prolonged gestation in all species of the subfamily Tenrecinae. The young mature quite rapidly after birth and reach a semi-independent status by one month of age. The larger T. ecaudatus matures slightly later than the smaller genera and species. Hemicentetes can breed in the season of its birth but the other species appear to delay breeding until the following year.
EISENBERG, J. F. and GOULD, E. (1967): The maintenance of tenrecoid insectivores in captivity. Int. Zoo Yb. 7: 194-196. (Free Full Text)
GOULD, E. (1965): Evidence for echolocation in the Tenrecidae of Madagascar. Proc. Am. phil. Soc. 109: 352-360.
GOULD, E. and EISENBERG, J. F. (1966): Notes on the biology of the Tenrecidae. J. Mammal. 47: 660-686. (Free Full Text)
Back to the top
Back to Tenrec Resources and Information