178 FLORIDA GEOLOGICAL SURVEY SIXTH ANNUAL REPORT. 



The writer has gradually evolved a method of quantitative analysis of 

 vegetation which, though necessarily less accurate than those just mentioned, 

 is much hetter adapted to rapid reconnoissance work. A description of it 

 follows : 



In traveling through Florida or any other part of the country where natural 

 vegetation is abundant, I try to jot down the name of every wild plant 

 observed, with some indication of whether it is abundant, common or rare, 

 repeating the notes at convenient intervals, usually at every milepost when 

 on a train, or at every stream or other topographic feature when walking. In 

 this way the largest and most abundant plants are Hkely to be noted oftenest, 

 which is just what is wanted in a quantitative study. Then at some subsequent 

 time I go through the field notes for each region (or even for each habitat 

 separately if desired) and count the number of times each species is men- 

 tioned, making allow^ance for relative abundance as follows. Where a species 

 has been noted as abundant I count it three times, where common I count it 

 twice, and where rare I do not count it at all. 



Even this does not do justice to the great abundance of the longleaf pine 

 and some other conifers, so I then usually multiply the figures for Pinus palus- 

 tris by 4 and those for P. Elliottii, P. Taeda, P. serotina, P. echinata, P. clausa, 

 Taxodinm imbricarium, and Chanvaecy parts by 2, and sometimes adjust the 

 figures a little arbitrarily besides, if the results are obviously inconsistent with 

 known facts. (It would probably be well to treat grasses in a similar manner, 

 but that has not been undertaken). The resulting figures can then be added 

 together and the percentages calculated. 



Allowance has been made for the difference in size between trees, shrubs, 

 herbs, etc., by the following device. Before calculating the percentages I take 

 the size of the average forest tree as unity, divide the figures for small trees 

 by 10, those for shrubs and woody vines by 100, and for herljs by icoo. If the 

 smaller flowerless plants were included, the mosses, lichens, and larger fungi 

 might be divided by 10,000, and so on "ad inftnitiim." This decimal system 

 is of course very arbitrary, but it has been adopted for convenience in dividing, 

 and it is probablj'' about as accurate as any other .system of equal simplicity 

 that could be selected. 



Sources of error. Of course it cannot be claimed that the percentages 

 obtained by this method are very accurate, but with all its imperfections it is 

 far superior to anything used for such a large area before, and future investi- 

 gations perhaps will not materially alter my main conclusions. The principal 

 possible sources of error in the present enumeration, and the checks on them, 

 are as follows : 



1. A great deal of the long-leaf pine, and some of the other trees, es- 

 pecially cypress, has been removed by lumbermen, so that these trees are rela- 

 tively less abundant now that they were originally. If it were desired to 

 re-construct the original condition of the forests, perhaps the use of 6 or 8 

 instead of 4 as a multiplier for long-leaf pine would give a pretty close ap- 

 proximation to the truth. 



2. The farmer in clearing land usually attacks the uplands first, making 

 most of the swamp plants relatively more abundant. But there are not many 

 parts of Florida where the proportion of cleared land makes much difference 

 yet, and this report seeks to represent present conditions, anyway. 



