Introduction
Eating an old shed skin (keratophagy) is very rare to see in snakes. KUCH (1998) reports, that only 19 cases of keratophagy in 14 snake species of the families Boidae, Colubridae, Elaphidae and Viperidae are certified. There exist different hypotheses about the biological meaning of this extraordinary behavior. For lack of data all of them are more or less speculative.
The Snake
The observations were made at an adult female of the Common Garter Snake Thamnophis sirtalis sirtalis "florida blue". The animal was a wild caught snake imported from the USA in 1997. During observation time this animal was kept single in a smaller sterile terrarium because of quarantine. Keeping conditions were similar to those earlier described (HALLMEN 1997).
Observation
On 28.4.1998 the snake was fed with 10 smelts (HALLMEN 1998). While swallowing one of the prey fish got into contact with an old shed skin the day before. From that moment the snake did not move any more during the act of swallowing. So the skin kept stuck to the fish.
First the animal did not seem to recognize that it began to swallow the skin with the end of the fish. After having swallowed the fish completely (it was in the esophagus 2-3 cm behind the head) the snake wanted to end the act of feeding. Than it became aware of the skin hanging out of its mouth. Immediately the garter snake began to move its head backwards to get rid of the skin.
That lasted about 3-4 minutes. Afterwards it stopped these unsuccessful efforts and swallowed the 15 cm long rest of the old skin hanging out of the mouth in the same way as it had done before with the prey fish.
The total length of the swallowed skin was about 30 cm. The complete act (1 smelt + the skin) lasted 20 minutes. I had to hurry to take a picture of this unusual event (figure 1).
3 days later the animal defecated. The excrement consisted of 2 bigger parts and was even for the defecation of a garter snake of a very strong smell. The skin swallowed 3 days before was clearly to be seen in the excrement (figure 2).
In the fresh defecation the skin could be extended to more than half of its original size. The skin was very elastic and didn't differ from a skin that had not been inside a garter snake for 3 days. I found no sign of digestion on it. In the 2 main hollow spaces formed by the skin there seemed to be undigested or not completely digested rests of the prey fish. I thought them to cause the extreme smell. I wondered how resistant the skin was to the aggressive gastric juice of a snake. To that day I had made out the gastric enzymes aggressive enough even to digest bones of smaller mammals.
Discussion
Following the summary of KUCH (1998) this is the first description of a keratophagy in garter snakes of the genus Thamnophis.
Based on my observation I think keratophagy at least in the genus Thamnophis is only the accidental swallowing of an object, that is or has been in contact with the food. Also other objects of the terrarium setting are swallowed with the food as I could observe for example with parts of the litter. According to this I can only think of 2 reasons for keratophagy in snakes of the genus Thamnophis up to now:
1. The skin is connected to the food (fish, mice or other). The snake does not make any difference between food or skin and so it is swallowed as "food".
2. The skin had intensive contact to the food and got its smell. The animal thinks of it as something to eat and swallows it.
Because the genus Thamnophis is very often kept in terraria, it is possible that there exist some unpublished observations about keratophagy. I would be very interested in any information about this theme.
Literature
HALLMEN, M. (1997): Some observations on stages of shedding in juvenile melanistic Common Garter Snakes Thamnophis sirtalis sirtalis. - The Garter Snake, 4/97: 19-23.
--- (1998): The Smelt Osmerus eperlanus as a prey fish for Garter Snakes of the genus Thamnophis in the terrarium. - The Garter Snake, 1/98: 24-28.
KUCH, U. (1998): Fälle von Keratographie bei einer Anden-Schlanknatter (Alsophis elegans), einem Bänderkrait (Bungarus fasciatus) und einem Vielbänderkrait (Bungarus multicinctus) mit einer Übersicht zur Keratographie bei Schlangen. - Salamandra, 34(1): 7-16.