74 JOHS. ANDERSEN 



their roe-deer and to replace them by others introduced from districts with 

 stocks of superior quahty. In all cases, despite the fact that our figures were 

 available as a guide, the number of animals actually killed far exceeded the 

 most optimistic estimate. 



It is noteworthy that population estimates in hare [Lepus euro pens) and red 

 deer [Cervus elaphtis), made in the same subjective way, tend to be equally far 

 from the mark. The human senses cannot replace more objective methods, 

 even when direct observation is supplemented by the study of abundant 

 snow tracks; we are therefore forced to conclude that simple observations 

 in the field will, invariably, lead to erroneous results when they serve as a 

 basis for population estimates of mammals — though perhaps to a smaller 

 extent when large and conspicuous animals living on open ground are being 

 counted. 



In 195 1 twenty-four roe-deer were introduced into the then empty woods, 

 and the herd was found to increase in size quite explosively. In 1955 we were 

 under the impression that the herd had reached its former size. Although the 

 entire herd could, theoretically, be descendants of the twenty-four deer 

 introduced, it is perhaps more reasonable to assume that a certain measure 

 of immigration had taken place. 



CENSUS BY MARKING AND RECAPTURE 



Since our work is primarily experimental we felt the temptation to analyse 

 in greater detail the size and composition of the fine roe-deer herd which 

 filled the woods to carrying capacity. After careful consideration, we decided 

 to do this by using a marking technique, rather than by shooting all the deer 

 again. 



Methods 



Although the principles used in capture-marking-recapture methods are 

 simple, the technique proved not quite so simple when applied to roe-deer 

 in practice. I shall here describe the method which we finally arrived at and 

 use now. 



The trap consists of a fenced enclosure 2 X 20 m with a gate at one end; 

 the gate can be closed by pulling a long wire. It is placed in a thicket in one 

 of the spruce plantings in which all our fences are situated. 



The trapping commences about Christmas; oat sheaves are placed in all 

 traps and the baiting is repeated two or three times a week throughout the 

 trapping period. The deer soon come to the enclosures for food, and during 

 an initial period they are allowed to enter and leave freely. 



The trapping proper starts in January, usually when snow covers the 



