214 TWENTY-FIRST REPORT. 



complexes by the now distinct forest types we may at least get a plausible 

 explanation of the problem. Without doubt certain successional stages ensued, 

 but it seems highly improbable that the full garnet of forest associations was 

 run in any case, not even in the maple-beech climax series, and in most cases 

 succession was decidedly abl)reviated or entirely lacking, as was probably the 

 case in the big pine association. In short, the original distribution of the 

 upland forest associations appears to have been more the result of preoccupa- 

 tion and subsequent self -perpetuation than of a well-marked successional devel- 

 opment. 



These conditions are well exemplified in Lake county. Within a radius 

 of some fifteen miles from Little Manistee are to be found five types of upland 

 forest associations — the big pines, the jack-pine, the jack -pine-oak, the oak 

 and the mixed hardwood. Observations would seem to indicate that luider 

 natural conditions each of these types is essentially self-perpetuating. The 

 only possible successional development vA^ould be that resulting in the jack- 

 pine and jack -pine-oak passing into the pure black-white oak formation. If 

 freed from the influence of fires and cutting my observations would seem to 

 indicate an almost interminable duration of the mixed jack-pine-oak associa- 

 tion for reseeding of the former species goes on abundantly. It is, however, 

 recognized that the jack-pine is a tree of shorter duration and more subject 

 to disease than the oaks, and this would of course work in the direction of 

 elimination of the former. Under artificial conditions the above postulated 

 succession is being rapidly accomplished and the jack pines will soon be a 

 thing of the past. Also under the influence of similar artificial factors the 

 big pines have a similar termination. It should also be noted that the jack 

 pines rarely exist in a pure stand of any extent, but rather as impure bands 

 or isolated islands among the oaks. On the other hand, I have never seen in 

 this region a pure stand of the oaks which was unquestionably the result of 

 natural successional development. 



An ecological classification should be based upon original conditions inso- 

 far as we may be able to reconstruct them. To this end the month of August. 

 1918, was spent in this region and an intensive study of soil and atmospheric 

 conditions was carried on. The results at hand fail to show any soil or 

 atmospheric conditions of sufficient gi-adient to explain present forest distri- 

 bution or to give any indication of casual successful factors. The following 

 table of moisture holding capacities of a series soils taken at a depth of 25 

 cm. and determined on the basis of equal volumes according to Hilgard's 

 method ('OG) shows clearly the inadequancy of the soil factor, as it reveals no 

 soil differences of casual magnitude. 



