APPENDIX 357 



The trapping of small animals has also been given attention. Townscnd 

 (1935) has worked out a plan for such trapping. The Illinois Natural His- 

 tory Survey has used a unique method in estimating fish populations. Num- 

 bers of fishes are caught and marked, and the proportion of these that enter 

 into subsequent catches is used to estimate total populations. For example, 

 if 10 per cent of the subsequent catches were made up of marked individuals, 

 it would be assumed that the number marked was 10 per cent of the total 

 population. 



The numbers of birds and larger mammals are usually ascertained by 

 cruising. Only experts are really effective in estimating numbers of birds. 

 The person generally best at field identification usually reports most birds. 

 The general impression is that most of the figures published are underesti- 

 mates. All observations should be made at the hour of maximum activity, 

 often early morning, but different for different species. 



Methods of preserving these animals in natural numbers in public reserva- 

 tion areas too small for their proper support are discussed by Hall (1929) 

 and Shelf ord (1933, 1936). This consists in surrounding the small sanctuary 

 area by buffer zones or zones of protection of animals traveling out of the 

 sanctuary in course of food-getting, seasonal migration, etc. Such areas are 

 of the greatest value to science. They constitute check areas for reservations 

 under management. 



In dealing with the more strikingly influent animals, for example, mam- 

 mals and birds, one is confronted with the fact that they have commonly been 

 exterminated or suppressed in most of the places where it is desirable to know 

 what their effects may have been under primeval conditions. In order either 

 to reconstruct a former condition or to determine an existing one in regard 

 to these animals, the following operations are necessary: 



1. The species of influent animals occurring in a biome should first be listed 

 for several points not too near the biome periphery. 



2. The range of these species in and out of the biome areas must be 

 ascertained as completely as possible. 



3. The relative abundance of a species in the parts of its range must be 

 ascertained from literature. The records are fragmentary and in obscure 

 publications. 



4. The breeding, shelter, and feeding preference must be determined in 

 relation to daily and seasonal cycles. Home range should be ascertained. 



5. These habitat preferences must be interpreted in terms of the various 

 serai stages and the climax. 



a. Water. 



6. Water margins. 



c. Bare areas of all kinds. 



d. Early serai stages. 



6. The distribution of the habitat type occupied or utilized by the animal 

 must be determined. Its local occurrence in other biomes as bare areas, devel- 



