30 BULLETIN 354, U. S. DEPARTMENT OF AGRICULTURE. 



broadleaf forest of the the north, is very rare. Decomposition is 

 extremely rapid under the influence of tropical heat and great humid- 

 ity, and these, together with more gradual leaf fall, extending over the 

 entire year, prevent the accumulation of litter. Then, too, the tor- 

 rential rains wash much of it off the steep slopes almost as rapidly as 

 it is formed. 



As to the trees themselves there is almost infinite assortment of 

 kinds, sizes, and forms. One of the most striking features is the large 

 number of light-colored, smooth-barked species resembling in appear- 

 ance our northern beech. 1 Then, too, the trunks of the trees forming 

 the main crown cover are very characteristic, being for the most part 

 of very unequal thickness, and usually more slender 2 than those in 

 the virgin forests of the Temperate Zone. Large trees up to 5 feet 

 in diameter above the root flare, however, are not lacking even to-day 

 in the Luquillo. There are,' besides, many trees, tabanuco (69) for 

 instance, with a much-buttressed base formed by planklike outgrowth 

 from the trunk and the uppermost roots. 



There is a striking lack of uniformity in association and in distribu- 

 tion of species. The reasons for this are the vast number of species, 3 

 the combination of accidental association that such a number makes 

 possible, and the absence of any considerable modifying soil or other 

 conditions tending to form fixed associations within the broader and 

 more uniform climatic one. 4 The presence or absence of a tree, par- 

 ticularly one of the more valuable kinds, like cedro, appears to be a 

 matter largely of chance. The really valuable trees seem almost 

 hopelessly in the minority, while the inferior species are so numerous 

 as to impress one with the apparent~worthlessness of the forest. Un- 

 questionably many of the so-called worthless woods are unjustly 



1 According to Schimper this is owing to the prejudicial effect of humidity on the formation of cork, the 

 bark thus remaining poorly developed. The formation of bark is often so poor that moderately large 

 trees show green, owing to the chlorophyll of the cortical layer being visible through it. There is, never- 

 theless, considerable individuality to the bark of different trees; some have thin flaky and scaly bark, as 

 in Myrtaceae, or a green surface, as in some Leguminosese; others, again, are armed with spines or corky 

 warts, while still others exude resins when wounded. 



2 This, according to Schimper, is a distinguishing characteristic of the virgin tropical forest. Woodward, 

 too, discussing the rain-forest in Santo Domingo, states that while trees over 5 feet in diameter and 100 feet 

 high are occasionally found, the average is far below these figures. 



8 Gifford and Barrett in their "Trees of the Luquillo Region" (appendix to Bulletin 54, Forest Service, 

 "The Luquillo Forest Reserve, Porto Rico") compiled a classified description of something over 100 identi- 

 fied species and enumerated besides the common names of nearly 100 more. 



* That the condition is not peculiar to Porto Rico, as many believe, and that, except in extent, the rain 

 forests of the Luquillo do not essentially dilfer from the other Antillean forests, the following will show: 

 Woodward remarks that in the virgin rain-forests of Santo Domingo two caoba(mahoganjO trees to the 

 acre constitute a good stand. Fernow, likewise, is speaking of the virgin forests of the Sierra Maestra, Cuba; 

 remarks that it was most puzzling to discover a law of distribution. "After many days cruising," he-says', 

 "over canyon, slope, and ridge one finds in identically the same kind of locality a new species, a single tree 

 or group never to be seen again in further cruisings. Nearly 400 miles had been traveled before the first 

 group of ebony was met." He further states that "the openness of the main stand may be judged from the 

 statement that as developed by some 1,200 acres of sample area, less than 1.4 trees of commercial size per 

 acre were found. When it is considered that over 100 species participate in making up this stand the diffi- 

 culties of a commercial or even a botanical survey will be realized." 



