134 



ANIMAL ECOLOGY 



shorter intervals of three or even two years. Now, by examin- 

 ing the records of skins obtained by the Hudson's Bay Company 

 in Canada, it is possible to find out when the lemming years 

 have occurred in Arctic Canada. This is made possible by 

 the fact that the arctic fox numbers depend mainly upon those 

 of lemmings, since the latter are the staple food of the fox. 

 The curve of fox skins, which shows violent and regular 

 fluctuations with a period of three or four years, can therefore 

 be used as an index of the state of the lemming population in 

 different years. When we compare the lemming years in 



Fig. II. — The fluctuations in numbers of lemmings are very violent 

 and very regular, and synchronise in Scandinavia and Canada. Curve A 

 shows the number of arctic fox skins taken annually by the Hudson's Bay 

 Company in Canada. Curve B shows the number of skins taken each year 

 in the whole of Canada, from 1 920-1 924. Diagram C shows the lemming 

 years in Canada deduced from the fox curves A and B. Diagrams D and E 

 show the known dates of maximum ("lemming years") of lemmings in 

 Norway and Greenland (the latter being incomplete). (From Elton. ^*) 



Canada and Norway the curious fact emerges that they 

 synchronise almost exactly in the two countries, and there is 

 little doubt that the numbers are controlled by some common 

 factor, which can only be climate. The other records which 

 exist show that Greenland and the islands of the Canadian 

 Arctic Archipelago also have lemming maxima at the same 

 time as the others. 



II. In the lemming we have therefore an animal which 

 undergoes regular and violent fluctuations in numbers from 

 year to year, which can be analysed into three main processes : 

 first of all, the epidemic which kills them off when a certain 



