Back to Tenrec Resources and Information

Care and breeding of the Madagascan hedgehog tenrec, Echinops telfairi, under laboratory conditions

H. Künzle

Summary

Insectivores are known to be difficult to breed, and breeding the erinaceous hedgehog under laboratory conditions is almost impossible. The present study describes how a large colony of lesser hedgehog tenrecs (Echinops telfairi) can be kept and bred under such conditions. These nocturnal insectivores of 100-150 g are kept in groups of two to three animals in relatively small cages. Walls with regularly spaced holes allow the animals to climb around. In addition, the tenrecs are exposed to reversed schedules of light, temperature and humidity in order to simulate the animals' annual cycle of activity in their native habitat, Madagascar. The meals consist essentially of canmeat, bananas and dryfood.

Following a gestation period of 50 to 60 days the hedgehog tenrecs give birth to an average of 3 to 4 young once a year. The number of parturitions relative to the stock of adult animals, is clearly larger in our colony than that reported from zoos, possibly due to our forced initiation of the mating phase. Unfortunately we have not succeeded so far to lower the tenrec's high mortality rate either in the weaning period, nor in adulthood (especially while awakening from their torpor). These deaths, however, are not considered to be due to an insufficient species-specific care, but to unknown causes that the animals are also confronted to in other institutions, in part even in the wild. Learning more about the care and breeding of hedgehog tenrecs is rewarding for the purpose of basic research as well as the preservation of the animals.

Zusammenfassung

Die vorliegende Arbeit beschreibt, wie der kleine Igel Tenrek, Echinops telfairi, (Tenrecinae, Insectivora) - im Gegensatz zum europäischen Igel - selbst unter erschwerten Laborverhältnissen gezüchtet werden kann. Die Tenreks (100-150 g KGW) werden in Gruppen von zwei bis drei Tieren in relativ kleinen Käfigen gehalten. Gitterartige Wände geben ihnen die Möglichkeit zum Herumklettern. Licht, Temperatur und Feuchtigkeit werden variiert, um Bedingungen zu simulieren, die halbwegs jenen in ihrem Ursprungsland Madagaskar entsprechen. Als Nahrung erhalten die Tenreks überwiegend Büchsenfleisch, Bananen und Trockenfutter.

Nach einer Trächtigkeitsdauer von 50-60 Tagen werfen die Tenreks einmal im Jahr durchschnittlich 3-4 Junge. Die Zahl der Geburten im Vergleich zur Gesamtzahl der Tiere ist gemäß International Zoo Yearbook bei uns deutlich größer als in den Zoos. Möglicherweise liegt dies daran, daß bei uns Weibchen und Männchen, nach dem Erwachen aus dem Trockenschlaf (Torpor), wiederholt ausgetauscht werden. Leider besteht nach wie vor eine relativ hohe Sterberate bei jungen und subadulten Tenreks. Da ähnlich hohe Raten aber auch von anderen Zuchten gemeldet werden und hohe Sterberaten auch bei freilebenden Igeln beschrieben sind, wird angenommen, daß diese Todesfälle nicht primär durch eine mangelnde artgerechte Haltung bedingt sind. Zuchtdaten von niederen Säugetieren dürfen nicht direkt mit jenen von Standardlabortieren verglichen werden. Die Haltung und Zucht der kleinen Igel Tenreks ist jedoch wichtig für deren Erforschung sowie zur Erhaltung der Art.

Key words

insectivores, tenrecinae, environmental conditions, life expectancy, mating, litter size, rearing, mortality rate

Introduction

The Madagascan lesser (pigmy) hedgehog tenrec, Echinops telfairi, may be less popular than the Tenrec ecaudatus, known to give rise to litters of up to thirty young (1, 2).

The phylogeny of insectivores (lipotyphla) is still a matter of debate (3, 4, 5). A useful grouping rather than a classification is obtained by subdividing the insectivores into three major populations the erinaceomorpha (including the european hedgehog), the soricomorpha (shrews and moles) and the tenrecomorpha. The latter group may be composed of the tenrecidae (tenrecinae and oryzorictinae), the chrysochloridae and the potamogalidae. Representatives of the tenrecinae are the lesser and the greater hedgehog tenrec, (Echinops and Setifer setosus respectively), the Hemicentetes semispinosus and the Tenrec ecaudatus.

It has been stressed repeatedly, however, that among the tenrecinae, the Echinops can probably be kept and bred the best, due in particular to its social behavior, its acceptance of an omnivorous diet and its life expectancy [(1, 6); as to the difficulties in breeding erinaceous hedgehogs under laboratory conditions see ref.7]. The maintenance of colonies of Echinops is important not only with respect to their preservation (8, 9), but also for the understanding of mammals in general, especially their phylogeny (10, 11), reproduction (12, 13), thermoregulation (14, 15) and neurobiology (6, 16). As to the latter, my own field of research, the Echinops is unique because its brain/body weight ratio and its neurocorticalisation index are among the lowest of all mammals (17), and unlike non-eutherian mammals, the Echinops does not show any degree of organized cerebral granularization suggestive of the presence of a cortical layer IV (18, 19). It is our objective to learn more about the neuronal organization of poorly differentiated brains, in order to get some insight into the variability and complexity of more differentiated brains.

It may be worth mention in this respect that we do not really know what's the significance of a poorly and a highly differentiated brain. The cortical layer IV e.g. is well developed and quite essential for the processing of ascending information in rodents, cats and primates; it is poorly differentiated, however, not only in insectivores, but also in the delphins showing a high brain body ratio and a convoluted cerebral cortex similar to man.

Finally we want to become aware of evolutionary trends, enabling us to extrapolate experimental findings from one species to the other, including man.

There are several reports dealing with the care and the breeding of the lesser hedgehog tenrec (1, 6, 9, 20-24). These studies, in addition, present details about the growth, development and behavior of Echinops, we do not consider here. The present study concentrates on the possibility of breeding Echinops in large colonies for the purpose of basic research.

Environmental conditions

Housing

Originally the animals were kept in large cages of 80 x 54 x 83 (high) cm. Due to our limited space, the increasing number of animals and the impression that this cage size may be unnecessarily large, we have gradually reduced the size of the cages to 45 x 60 x 34 cm. The aluminium cages designed originally for rabbits have been altered in the way that 2-3 walls (in part also the ceiling) were made of (or contained inside) a stanced metal plate with regularly spaced holes of 8-10 mm² (Figs. 1, 3). These walls have proved to be more stimulating for the animal to climb around than branches of trees, which were used only for a short time after occupancy. Except for breeding purposes (see below), the cages are used for up to six animals of the same sex. The cage floors are lined with dust-free sawdust. Depending on the activity phase of the animals the cages are cleaned every second to sixth week (weekly in case of large litters).

Each cage also contains at least one wooden nestbox of an internal height of 12 cm (Figs. 1, 2). The box consists of an anteroom (7 x 10 cm) and a main room (10 x 10 cm) both covered with a lid which can be easily lifted (23). Within the main room we have seen up to four adult animals packed tightly together, even in cases where there was an additional empty nestbox in the cage.

Light, temperature, humidity

In order to simulate the annual cycle in activity of their native habitat in south-western Madagascar (high aridity, relatively low temperature and scarcity of food during the torpor phase) the tenrecs were exhibited to reverse schedules of light, temperature and humidity similar to those described [(23), Tab. 1]. We agree with these authors that absolute synchronisation between individuals is scarcely possible to obtain. Furthermore the time where most animals fall into the torpor phase, or come out of it, varies within weeks from one year to the other. A relatively distinct consistency has only been obtained in regard to the delivery time (see below), probably due to the forced initiation of the mating phase.

Our animals are kept in two rooms, a main room regulated as described and a small poorly regulated room. In the latter room we keep the animals considered not useful for breeding purposes (including runts and old animals). Nevertheless, parturitions are also observed in this room and it even occurs that the bad condition of an animal improves following its transfer from the well-regulated to the poorly regulated room.

Ultraviolet illumination (in addition to ordinary fluorescent light) had been given for three hours a day (following feeding) until spring 1992. Thereafter UV-illumination was stopped because of the relocation of the animals into another room due to reconstruction work, so far without obvious harmful effects.

Feeding

The animals' food consists of either (I) canned cat food containing beef of fish in an alternate fashion, (II) dog food (sausage), (III) bananas or (IV) dry food, composed of hedgehog food (Vitakraft, Bremen) and dry food for cats (e.g. brekkies).

The nutrition is given in the early morning according to the schedule in Tab. 1. Water is always present in the cage and renewed twice (in torpor phase) to 5 times (activity phase) a week. The favorite meal, undoubtedly, is the can food with beef, while the dry food is the least appreciated. A meal of dry food, however, is convenient to serve over the weekend. In addition, the brekkies force the tenrecs to bite and the hedgehog food (only eaten if the tenrecs are hungry) helps to prevent over-feeding of the animals.

Time of change	Phase	Light hrs	Temp. °C	Humi- dity %	daily meals
					mo	tu	we	th	fr	wk
mid Nov	torpor	2330-1000 (10.5)	20-21	40-50	a	--	--	--	d	--
end Jan	partial awaking	2300-1000 (11)	22	"	a	--	b	--	d	--
end Feb	main awaking	2230-1030 (12)	24	50-60	a	--	b	--	d	--
end Mar	mating	2300-1200 (13)	24-26	60-70	ag	b+d	--	a	d	--
end Apr	pregnancy	"	"	70-80	a	b+d	--	a	d	--
end June	rearing	2330-1200 (12.5)	"	"	a	b	b+o*	o	d	b+d*
mid Aug	resting growth*	2400-1130 (11.5)	22-24	60-70	a	a*	b	o	d	a*
end Sept	pretorp growth*	2330-1000 (10.5)	"	50-60	a	a*	b	a*	d	--

Tab. 1:

Light, temperature, humidity and feeding schedules a, canned cat food containing either beef or fish; b, bananas; d, dry food (hedgehog food supplemented with brekkies); g, chopped garlic; o, dog food (sausage); wk, weekend.

* applies to young animals only; light is provided by ordinary fluorenscence lamps; water is given ad libitium.

Initially we also gave mealworms twice a week during the activity phase. This feeding procedure, however, turned out to be too time consuming for large colonies. Furthermore, mealworms were eaten predominantly by the heaviest animals, and there were often mealworms lost and subsequently turned into meal-beetles, which our tenrecs didn't try to catch. The tenrecs, on the other hand, like the chopped garlic given once a week during the mating phase.

Life expectancy of adults

Because our tenrecs are used ultimately for research we have no precise data regarding their life expectancy. Nevertheless, there were at least 40 animals reaching an age of six years or more, one female gave birth to six young (five of which survived) at an age of eight years, and the oldest Echinops died at the age of eleven years.

Unfortunately there is quite a high mortality rate among our tenrecs! The annual rate of adult death is about 15-20 % (7-29 %) of the animal stock. Except for the mite epidemy we had this year, the causes of death are largely unknown. Only a few of our tenrecs die of old-age (old animals are often perfused for immunohistochemical studies), while some others die due to infections, particularly mouth-jaw abscesses. Among the animals dying without obvious causes, the majority of deaths occurs between January and April and may be considered physiological events due to the difficulties in awaking from torpor [It may be mentioned in this respect that for adult/subadult wild erinaceous hedgehogs there are estimates of winter losses ranging between 26 to 43 percent (7).]. It is also notable, that roughly half of the animals dying from unknown reasons die at an age of one or two years, and most of these appeared not to be sexually active.

Mating phase

In view of the fact that Echinops is giving birth only once a year, and the final litter size is relatively small, it is important, of course, to get a mating rate as high as possible. While there were tenrecs mating at a time when we still considered them being in the late torpor phase, most females are mated in April or later (light, temperature and humidity parameters have been raised long before; see Tab. 1). This fact has forced us to stimulate more rigorously the mating process by a regular exchange of females (All animals used for breeding purposes are marked and recorded; subsequently, attention is given that animals with identical parents and grandparents can not mate.). Thus during April and early May, every 8-10 days, the females are taken out from the cage and placed with another male in a different cage. Attention is also given that parous and non-parous females get into contact with proven and unproven males. It is possible to keep two females with one male particularly if the latter are limited. In this case, however, females are not exchanged simultaneously but successively within two or three days. This exchange procedure increases obviously the rate of successful mating (the dates of birth often correlate closely with the dates when females are put with another male) and limits the delivering phase to roughly one month (actually June instead of June to mid-August when this procedure has not been applied. The latter effect extends the time for the young to grow up after being weaned (from September onwards adult tenrecs already start to enter the pre-torpor phase), and another advantage is that it increases the possibility of fostering a litter by a foreign mother. A regular exchange of tenrecs, furthermore, often helps to identify the animals's sex (see legend of Fig. 1) and thereby avoids unnecessarily long "pairing" of tenrecs of similar sex.
Despite this exchange procedure there remained, however, some inconsistencies. Thus we have noted repeatedly some primipara at an age of 3 or 4 years. In addition, there were two females of 5 and 6 years which delivered healthy young a second time, following an interval of three years. Interestingly, we obtained and gave away so-called unsuccesful pairs of Echinops, which reproduced successfully as soon as they were loged in the new environment.

Pregnancy and birth

We recorded repeatedly gestation lengths of less than 60 days, and in one case, the pregnancy lasted clearly 52 days or less (for comparison the european hedgehog has a gestation length of about 35 days). Thus the gestation length of our Echinops is less than the 62-65 days usually referred to (1, 2) but similar to the data reported by FOWLER 1986 (thesis cited in ref. 25). The differences may be explained by the different environmental conditions used, particularly our relatively high temperature (25).
In our colony 741 parturitions, with 2458 newborn, are recorded so far (Tab. 2).

animal Year	litter stock a	young number	litter ini b	young size	elid Nov c	death rates li/yg d
							ini e	adj f
1980	48	12	48	4.0	25	2/10	48	34 (1)
1981	88	1	1	1.0	1
1982	59	1	2	2.0	1
1983	48	8	18	2.3	15	0	17	17
1984	61	13	34	2.6	31	0	9	9
1985	97	18	57	3.2	38	3/11	35	17
1986	121	22	78	3.5	46	2/9	41	33 (2)
1987	162	26	108	4.2	78	2/10	28	20
1988	231	36	151	4.7	128	0	12	12
1989	281	68	216	3.2	141	9/31	35	24
1990	325	75	232	3.1	152	5/19	34	28
1991	307	47	154	3.3	89	8/34	42	26
1992	358	74	271	3.7	189	9/40	30	18 (3)
1993	358	77	252	3.3	166	8/31	34	25
1994	363	85	270	3.2	191	6/22	29	23
1995	424	58	190	3.3	142	5/26	25	13
1996	407	68	231	3.4	156	3/12	32	29
1997	429	52	145	2.8	62	9/37	57	43 (4)
total		741	2458	3.3	1651		33	24

Tab. 2:

Litters, newborn and their life expectancy This colony originates from 21 female Echinops (with a total of 43 parturitions) received between 1979 and 1982 (a) number of adults/subadults registered mid-March [except 1980, see (1)]; (b) initial number of newborn; (c) newborn surviving until end of November [early October respectively in 1997]; (d) so-called entire litter deaths (elid) with li referring to the number of litters with 3 and more newborns all of which died within 4 to 6 wks after birth; yg indicates the total number of newborn involved in this entire litter death; (e) true death rate in percent 5 months after birth; (f) adjusted death rate in which entire litter death are not taken into consideration. Special remarks: (1) most females received as pregnant; (2) absence of author during the rearing phase and dislocation of the animals to Munich in fall; (3) dislocation in a new room in spring (4) mite epidemy during late pregnancy and rearing.

This corresponds to an average litter size of 3.3. The two largest litters were of 8 and 9 young, from which, however, only 3 animals survived. A much better survival rate was obtained from the litters consisting of 7 (n = 12), 6 (n = 38) and 5 (n = 102) young. Older females tended to have slightly larger litters than the on year olds (Tab. 3). Similarly the average litter size of primipara (females of usually one or two years of age) was slightly smaller than those of multipara. In these parturitions 360 females were involved, 67 of which gave rise to 10 or more young during their life (14 females with 15-20 young; 6 females with more than 20 young). The best mothers were Et 253 (a total of 29 young in 7 parturitions), Et 293 (25 young/6 parturitions) and Et 540 (25 young/4 parturitions).

	mother age	litter number	young number	litter size	death rates
					ini %	adj %
1992
	>=4	18	73	4.1	16	10
	3	21	76	3.6	26	19
	2	15	45	3.0	49	30
	1	7	22	3.1	59	31
	pp	25	75	3.0	51	31
1993
	>=4	21	88	4.2	22	15
	3	6	24	4.0	66	40
	2	13	41	3.2	56	23 (1)
	1	30	80	2.7	68	25
	pp	44	128	2.9	37	23
1994
	>=4	16	45	2.8	40	21
	3	11	43	3.9	33	33
	2	35	117	3.3	25	19
	1	21	58	2.8	33	29
	pp	41	121	3.0	25	17
1995
	>=4	5	17	3.4	18	18
	3	24	91	3.8	38	19
	2	21	61	2.9	13	13
	1	6	17	2.8	24	0
	pp	23	64	2.8	19	13
1996
	>=4	15	55	3.7	22	16
	3	28	98	3.5	32	29
	2	11	35	3.2	23	23
	1	11	31	2.8	39	27
	pp	27	75	2.8	40	33

Tab. 3:

Litters, newborn and death rates in relation to the age of the mother; adj, adjusted; ini, initial; pp, primipara; (1), including the loss of one litter of 9 young

Rearing

Cages with litters contain at least two nestboxes (Fig. 2). Due to lack of space most litters are reared in the presence of the males. Clear evidence that males influence negatively the rearing of the young has not been obtained. Males, however, may severely attack the young entering the cage from a neighboring cage (this has occured particularly at the time when we subdivided large cages by a partition wall).

The general growth and development of our young correlate well with what has been previously described [(1, 21, 23), see also legend of Fig. 2]. We are, however, confronted with a relatively high mortality rate: namely 33 % within the first five months following birth (Tab. 2). This rate also includes the deaths due to the mite epidemy that we had this year, during the late pregnancy phase and the early rearing phase (This loss may partly be due to premature births, partly to the discomfort and stress the females obviously had.). We also consistently encounter a few deaths of entire litters (see elid in Tab. 2) due to postparturitional complications (death of dam or major neglect of newborn by mothers). If we exclude from our statistics all cases of entire litter death (litters with at least three young all of which die within six weeks following birth as e.g. the litter of nine young mentioned above, but not a litter of eight young three of which survived, or a litter of two both of which die) the mortality rate declines to 24 % (so-called adjusted death rate in Tabs. 2, 3). Apart from some special events (Tab. 2, remarks 1, 2 and 4) we did not succeed in finding clear causes of death for this remaining mortality. Unlike the entire litter deaths the age of the mothers does not appear to have an influence on this rate (Tab. 3).

It should be mentioned that an additional 7-8 % of the young may die during the following torpor phase; some young never gain normal size (runts); and there are others which just remain sexually inactive.

Concluding remarks

People engaged in the protection of animals may well be worried about the colony's high mortality rate - a possible indicator for insufficient, i.e. non-species specific handling of animals. Such a suspicion also arises if we compare our data with those of EISENBERG (1), HONEGGER (21) and collaborators - pioneers in the setting up of efficient breeding colonies of tenrecs. However, comparing our data with larger colonies and across a more extensive time period than those mentioned, our mortality rate my be acceptable. It is close to the one reported by GODFREY and OLIVER (23: twelve years record with 32 parturitions and a mortality rate of 28 %) and may even be better than that in zoos (Tab. 4). This suggestion is based on the fact that out of the 568 Echinops born worldwide in the zoos between 1980 and 1994, only 385 animals have been specified in regard to their survival and this selected population of animals already shows a mortality rate of 36 % (Tab. 4). Also, with respect to infant-mortality in the zoos, the length of their survival time is usually defined as 30 days, while our survival time is five months. Last, but not least, it may also be mentioned that the erinaceous hedgehog has a preweaning mortality of more than 40 % in captivity and nearly 20 % in the wild (7, 26, 27).

It would be ideal, of course, to lower these mortality rates. Not knowing, however, how to reach this goal, the establishment of an efficient breeding colony depends essentially on the increase of the reproduction rate, especially the birth rate1 (The litter size, of course, also influences the reproduction. In fact, this size is relatively low in our colony [3.3 as compared to 4.2 in the Jersey colony (23)], in part possibly due to the relatively high number of one-year mothers obtained by our mating procedure (Tab. 3).). Our data are exceptional in this respect. Thus, while in the last five years (1990-1994) the zoos' average annual number of 110 adult and subadult tenrecs resulted in an annual average of 22.4 young (data taken from the International Zoo Yearbooks), the respective numbers in our colony are 342 and 235. In other words, while the average annual numbers of adult tenrecs in our colony (342) exceeds the one in the zoos (110) by a factor three, the average number of our newborn (235) exceeds the one in the zoos (22) by about a factor 10 (Tab. 4).

Year	Adults tot/col	newborn			represent. colonies (1)
		tot/col	spe/col	dd	Br	Fk	Jy	Ph	Tp	Wa
1980	130/16	33/6	24/2	11	x	x	16+	x	8+	x
1981	82/14	66/4	66/4	24	x	21+	18+	x	24+	x
1982	88/19	72/4	69/3	21	x	20+	25+	x	24+	x
1983	113/21	57/7	39/2	17	x	4-	7+	4-	32+	4-
1984	89/19	84/11	56/6	10	x	13+	1+	5-	31+	10-
1985	109/20	57/8	26/3	9	x	5+	x	4-	19-	14+
1986	154/22	13/3	4/1	1	x	4+	x	7-	x	2+
1987	137/23	9/4	7/3	2	x	1+	x	x	4+	x
1988	111/26	34/5	23/3	11	x	x	x	5-	9+	6+
1989	103/23	31/6	20/3	7	x	x	x	4-	6-	7+
1990	114/27	36/8	19/4	9	3+	x	x	6+	9+	5-
1991	126/30	15/3	6/1	4	6+	x	x	x	x	x
1992	93/27	27/5	9/1	6	6+	x	x	x	7-	x
1993	94/30	25/5	15/2	5	3-	x	x	x	9+	4-
1994	127/32	9/3	2/1	2	2+	x	x	x	6-	1-
total		568	385	139

Tab. 4:

Data from other institutions keeping Echinops according to the Int. Zoo Yearbook Vol. 22-35. (1) a representative colony is defined as an institution/zoo reporting the birth of Echinops over a period of at least five years between 1980 and 1994; the numbers refering to these colonies indicate the total numbers of newborns; +/- after the numbers indicates whether or not the newborn were specified in regard to their survival; x indicates no newborn reported. Br, NY Bronx USA; Fk, Frankfurt Ge; Jy, Jersey GB; Ph, Philadelphia USA; Tp, Topeka USA, Wa, Washington NZP USA. Further abbrevations: tot, total number of adults/newborns; col, number of colonies involved; spe, animals specified in regard to a survival of at least 30 days; dd, number of specified animals which died.

It is highly unlikely that we keep our tenrecs in a more species specific way than zoos, but it is possible that our rigorous mating procedure helped to increase the number of parturitions. An important factor for a longterm rearing of Echinops, furthermore, might be the presence of a large stock of animals. There are great variations in both the number of newborn and adult deaths from one year to the other. We do not know the reasons for these variations, but one may imagine that without a large stock of animals a colony may suddenly be unable to breed. In any case, we have no other explanation for the sudden failure of several representative colonies to reproduce (Tab. 4).

Insectivores are certainly more difficult to breed than standard laboratory animals. Breeding the lesser hedgehog tenrec, however, is rewarding for basic research as well as for the preservation of the animals, particularly in view of the doubtful perspectives they have in Madagascar (8, 28).

Acknowledgements

The author deeply appreciates the invaluable help of S. SCHALLER, A. NEKIC and A. ANTONIUS for the care and routine husbandry of the hedgehog tenrecs. He is indepted to R. E. HONEGGER (Zürich) for his encouragement and help in setting up the tenrec colony. Thanks for advice and help are also extended to O. APPERT (Manja), Dr. K. BÜTTNER-ENNEVER (Munich), R. DMOCH (Frankfurt), W. L. R. OLIVER (New Jersey), Dr. R. VOARA (Tananarive) and Dr. R. WINKLER (Basel).

The maintenance of the colony and the research is supported by the Deutsche Forschungsgemeinschaft (grants Ku 624/1 and 624/2-1).

References