ANIMAL INTER-RELATIONS 29 



When we come to consider the predatory forms 

 we can discern other reasons why the number of 

 dififerent species is Umited. Each herbivorous animal 

 is usually preyed on by a good many predators, but 

 the latter do not usually restrict themselves to one 

 host only. Longstaff (1932) found that Passerine 

 birds in West Greenland (such as the Lapland and 

 snow buntings and wheatear) showed a very wide 

 range of choice in their food. Pentelow (1932) has 

 shown the same thing in the case of the brown trout 

 in England. The trout studied by him appeared to 

 eat practically any form of small animal available 

 in the habitat. One of the chief factors limiting the 

 choice of food in these polyphagous forms appears 

 to be size. Thus the trout only ate young crayfish 

 and not older ones. Sticklebacks eat young Gam- 

 marus pulex, but only up to the second or third moult 

 stage. An Arctic fox will attack a ptarmigan and 

 eat it ; it will attack a pink-footed goose, but only 

 to try and drive it away from the eggs. The Arctic 

 hare it cannot destroy. This sort of relation between 

 the sizes of predators and the animals they prey upon 

 is very important in splitting up the animal com- 

 munity into food-niches. The limitations placed by 

 size of food on feeding together with other special 

 food-preferences give rise to food-chains, leading 

 usually from smaller to larger forms, and starting 

 of course from some herbivorous or scavenging form, 

 which in turn depends directly or indirectly on plants. 

 An example of a Sub-arctic food-chain is a springtail 

 (pollen-feeder or scavenger) eaten by a hunting spider 

 which is in turn eaten by a Lapland bunting. On 

 an English heather moor the species would be different 

 but the niches the same, e.g., the bird would be a 

 meadow pipit instead of a Lapland bunting. An 

 example on pine wood country in Surrey or Hampshire 

 is the small fly that is caught by a small spider, the 

 latter by wood-ants, the wood-ants by a green wood- 

 pecker, and the latter possibly by a sparrowhawk. 



