This paper was written to defend the ethological concepts of fixed action pattern (FAP) and innate releasing mechanism (IRM) against criticism that they encouraged an artificial distinction between innate and acquired behaviours. Two main blocks of evidence were reviewed.

The first block of evidence had to do with nesting behaviour in the rat. This is a genuinely innate behaviour, because it takes place in rats reared in isolation and therefore unable to have acquired it by learning. In many ground-dwelling rodent species, this takes place in two major phases, namely (1) digging the burrow, and (2) nest construction (although the second phase actually involves repeated cycles of foraging for material and putting said material in place). Here are the sub-behaviours associated with each phase .....

Nevertheless, the nest is only built successfully if the above skills are properly triggered and coordinated. There has to be a suitable nesting site, for example, and it has to be suitably peaceful. So instead of arguing whether a behaviour is innate or not, it is better to regard it as made up of smaller behaviours, certain of which "are completely independent of learning" (p58). It is then the most effective sequencing of these behaviours which is learned.

The second block of evidence had to do with the killing technique of the polecat, and was an extension of earlier work by Kuo (1930). Kuo had raised 20 cats in isolation from weaning, and compared their predation behaviour with 21 cats raised with their mothers and 18 cats raised with a rat as a litter mate. When tested at around 4 months of age, 18/21 of the normally reared cats (86%) successfully killed a rat presented as prey. Of the isolates group, however, only 9/20 (45%) killed their prey, and with the rat-friendly group only 3/18 (16%) killed. Indeed, the rat-friendly group showed "peaceful and positive" reactions to the rats, licking and defending them and looking for them if they were temporarily removed.

Eibl-Eibesfeldt then presents his main report. He raised 20 polecats under conditions of experiential deprivation, specifically lack of experience with living prey (they were fed either prepared meat or dead rats). Eight animals were raised normally, and the remaining 12 were isolated from age 21 days. Eibl-Eibesfeldt then tested the isolates with live prey. Six isolates were tested at age 5 months, one at 10 months, and five at two years. Here is how he described the behaviour of the 5 month group: 

The point is that the nearer the head the polecat manages to grip, the harder it is for the rat to turn to defend itself. So the killing bite is actually doubly useful - not only is it inherently more efficient, but it is inherently safer too Eibl-Eibesfeldt reports that the polecats learned this grip quite rapidly. One animal killed within 20 seconds with only three attacks, and the others needed 1-15 minutes. After 4-6 kills, one per day, all polecats were skilled hunters and their killing bites were always directed immediately towards the prey's neck. So again we see adult behaviour as being based upon a number of innate FAPs, roughly sequenced and clumsily delivered to start with, but capable of considerable improvement with practice.

The report ends with the moving tale of how a big bad polecat refused to eat his friend the little chick, despite being ravenously hungry. Five polecats were fed dead chicks and then at 2 years of age were exposed to live chicks. Once again, there was no attack behaviour unless provided the chick stayed still ..... 

Eibl-Eibesfeldt then reduced these two polecats' daily feed, and offered them the chicks each day as their hunger increased. For the next four days the behaviour towards the chicks did not change. They carried the chick back to the nesting box, muttering readiness for social contact. Eibl-Eibesfeldt then left the chicks with the polecats overnight, and on the morning of the fifth day polecat B had killed and partly eaten its chick. When it was given another it was carried alive back to the next, but then killed and eaten half an hour later. Polecat A, however, still refused to attack its chick even on the 6th day, but when Eibl-Eibesfeldt killed the chick himself it started eating it immediately. 

Eibl-Eibesfeldt, I. (1961). The interactions of unlearned behaviour patterns and learning in mammals. In Delafresnaye, J.F. (Ed.). Brain mechanisms and learning. Oxford: Blackwell.

Kuo, Z.Y. (1930). Genesis of the cat's behaviour towards the rat. Journal of Comparative Psychology, 11, 1-35.