478 



THE COMMUNITY 



ecotone into grassland and ecotone into 

 forest, of such important influences as light 

 intensity, air temperature, soil temperature, 

 relative humidity, wind velocity, and rate 

 of evaporation. There are less easily meas- 

 ured gradients in erosion and humus com- 

 position. The windbreak efiEect of the for- 

 est-grassland ecotone, operating with the 

 wind velocity factor, estabHshes an easily 

 demonstrable horizontal gradient in num- 

 bers of wind-dispersed plants. There is also 

 a horizontal gradient in wind-blown drift. 

 This gradient in wind-dispersed plants is 

 naturally different from the horizontal gra- 

 dient in bird-dispersed plants. 



Studies on forest edge birds have been 

 reported by Beecher (1942), Carpenter 

 (1935), and Van Deventer (1936), among 

 others. The last-named carried out observa- 

 tions on winter birds in New York where 

 an extensive ecotone existed between up- 

 land forest and swampy woodland. In this 

 study particular attention was paid to the 

 tree sparrow, black-capped chickadee, 

 downy woodpecker, and white-breasted 

 nuthatch. All four species selected primar- 

 ily the swamp-forest edge during the period 

 of study between December, 1934, and 

 February, 1935. Van Deventer estimated 

 the probability of chance association of any 

 two of the four species at approximately 

 5 per cent, calculated by ascertaining the 

 percentage of the total bird population of 

 the area constituted by each species, then 

 averaging all possible combinations of these 

 percentages. Two or more of these four 

 species were associated on 31 per cent of 

 all observations. This gave an actual per- 

 centage of association approximately six 

 times as great as association resulting from 

 pure chance. This is statistical confirmation 

 of the general belief that ecotones, while 

 not existing apart from the parent com- 

 munities, still have a characteristic biota. 



The importance of the ecotone has been 

 realized by Kendeigh (1944) in a discus- 

 sion of bird populations. He calls attention 

 to the necessity of separating forest-edge 

 birds from forest-interior birds in comput- 

 ing population densities. 



The earlier seasonal development of the 

 ecotone, its abundance of habitat niches, 

 and the distinctive variety of foods avail- 

 able, as well as the relative security from 

 immediate pursuit, combine with the 

 twenty-four hour and seasonal intermediate 



character of its physical environment to 

 maintain a distinctive biota. 



The ecotone biota in general has received 

 Uttle study. One of the few papers is that 

 of Carpenter (1935), studying areas of 

 community competition in central lUinois, 

 where the ecotone biota had a structure in- 

 termediate between meadow and forest, 

 strongly aflPected by seasonal factors, owing 

 to its exposed surface, and with a sufficient 

 number of ecotone species to give such 

 margins a characteristic habitus. It should 

 be remembered that there is a general ten- 

 dency for ecotone animals and plants to 

 persist along roadsides and fence rows for 

 a certain length of time after the original 

 communities have been destroyed. 



The prairie-forest ecotone owes its com- 

 position to (1) statistically significant frac- 

 tions of species populations found in grass- 

 land or forest, or both communities; (2) 

 stray or incidental species present for tem- 

 porary protection; and (3) ecotone species 

 present by preferential selection. 



STRATIFICATION IN FOREST 

 COMMUNITIES 



At a distance, the mature forest appears 

 as a distinctly bulky, unorganized mass of 

 vegetation. At first one is apt to be im- 

 pressed by this mass rather than by the 

 organization of the community. Such first 

 impressions are succeeded by a reaUzation 

 that the forest consists of many Hving units, 

 the most obvious of which are trees. 



Further consideration suggests at least 

 three important generahzations. First, sheer 

 bulk of the forest changes the physical en- 

 vironment so that those physical influences 

 acting upon the forest are themselves 

 modified, and the forest consequently tends 

 to have a characteristic cUmate. 



Next, the arrangement of the forest or- 

 ganisms is not haphazard, but tends to be 

 orderly throughout. This is more obvious 

 in plants (Gleason, 1936), but not neces- 

 sarily less true of animals, although, as a 

 consequence of their activity, they are less 

 obviously integrated. As has already been 

 shown, a patent feature of this orderly ar- 

 rangement of organisms within the com- 

 munity is its stratification. Since each 

 stratum (lamiation of Carpenter, 1938) oc- 

 cupies a definite horizontal or vertical por- 

 tion of the community, the general forest 

 climate is subject to stratal modification. 



