A kept brood fed not on yolk, not through a wall, but on the keeper herself — a caecilian mother whose outer skin thickens and enriches, and whose young peel it off in strips with specialised teeth and eat it, while she stays, spent and alive
R902 closed with a mother who wore the keeping on her back and then shed the room: a Surinam toad whose skin sealed each young into its own pocket, sheltered it on its yolk for a season, and was discarded once the toadlets had pushed out through it. Her young never ate any part of her. Its closing turn asked for the exact reversal in that one respect — a keeper whose skin is not the room but the food. This page takes her directly: the Taita Hills caecilian, Boulengerula taitana (spelled taitanus in some older literature), a legless, ringed, burrowing amphibian endemic to the Taita Hills of southeast Kenya, in the family Herpelidae.
Eggs in earth, young born unfinished. The species is egg-laying and direct-developing: there is no free aquatic larval stage. The female lays her eggs in a chamber she makes in the soil, and stays with them. Broods are small — two to nine young in the accounts used here, described as among the smallest clutches known for caecilians. The hatchlings emerge altricial: about 28 mm long, pale, weakly muscled, with skeletons still poorly ossified, in no condition to dig or hunt for themselves. What they are born equipped with, ahead of almost everything else, is a set of specialised, fetal-like teeth — teeth whose whole first job is not prey.
The skin becomes a meal before it is taken. A brooding mother looks different from other adults: noticeably paler. The change is structural, not cosmetic. In a nursing female the outermost layer of the epidermis is transformed — its normally flattened, keratinised cells become elongated and filled with vesicles, rich in lipids and protein, and the epidermis as a whole grows to about twice the thickness found in non-brooding females. Read that plainly against the last two keepers. The gastric brooder’s young changed her organ’s chemistry to make it safe to be inside. The Surinam toad’s skin grew over her young to make a room. This mother’s skin grows rich in advance of being eaten: the keeper prepares herself, in her own surface, into the brood’s food supply.
Peeled, not nursed. The young position themselves over different parts of their mother’s body and work their lower jaws under the outer layer, lifting and peeling strips of it away and swallowing them. The behaviour has a name — maternal dermatophagy, skin-eating — and it was first reported for this species in 2006, from laboratory observations of mothers kept with their broods in soil, including video of the feeding. It is not nursing by secretion and it is not the young grazing on slime or shed skin found lying about: the food is living tissue made for the purpose, taken directly off the keeper’s body by the kept themselves. Between bouts the outer layer is renewed, so the feeding can repeat — she is not peeled once, down to something raw and finished, but eaten in layers over the whole period of care.
Fast growth, a measurable price. In the first week of skin-feeding in the original observations, the young grew at about a millimetre a day — roughly an eleventh added to their length — on skin alone. The mother, over that same week, lost about a seventh of her body weight, around fourteen per cent. Longer accounts agree on the direction even where they differ on the term: brooding females carry less body fat, and their body condition declines significantly across the care period. How long the dependence lasts is kept here as the accounts leave it — up to about four weeks in one report from the original team, with other summaries describing the young as reliant on her for their first two months — before the specialised teeth give way to the adult dentition, the young reach independence at roughly 86 mm, and they become what their mother is: generalist predators of termites, earthworms, ants and other soil invertebrates.
Why this keeper is different from the others in the run. Every keeper so far has paid in a currency that could be named apart from her body’s substance: a voice suspended, a stomach switched off, a layer of skin shed after it had served as a wall. This mother pays in the substance itself, converted in advance into food and removed from her, bite by bite, by the brood it feeds. And yet she is not consumed. The layer regrows, the bouts repeat, and at the end of the keeping she is still there — measurably lighter, measurably poorer in fat, alive in the same soil. The species, for all the singularity of its care, is listed as endangered, its forest fragments in the Taita Hills under pressure from deforestation and the intensification of the low-intensity agriculture it can otherwise tolerate.
One brooding, in order: a female lays a small clutch in a soil chamber and stays with it, her outer skin thickens to about twice its usual depth and fills with lipid- and protein-rich cells until she pales, altricial young hatch with specialised teeth and peel strips of that prepared skin off her body with their lower jaws and eat them, the layer renews between bouts while the young grow about a millimetre a day and she loses about a seventh of her weight in the first week, and the young leave her only when their adult teeth and bodies can take over. Schematic, compressed from the published record.
Walk it in order.
Schematic only: the stages compress work across accounts into one brooding. Labelled cautions: the growth and weight-loss figures are from the original laboratory observations over the first week, not a rate guaranteed for every brood or every week of care; how long skin-feeding dependence lasts is not settled to one term — about four weeks in one account, the first two months in another — and neither is used here as exact; the claim that the outer layer renews between bouts is the general account of the system, not a timed interval measured for this species on this page; and dermatophagy in this species is peeling and eating the mother’s prepared outer skin, not drinking a secretion and not scavenging shed skin.