Front Cover
 Part 2: Plant ecology of Porto...

Title: Scientific survey of Porto Rico and the Virgin Islands
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00091487/00009
 Material Information
Title: Scientific survey of Porto Rico and the Virgin Islands
Alternate Title: Scientific survey of Puerto Rico and the Virgin Islands
Physical Description: 19 v. : ill. ; 24 cm.
Language: English
Creator: New York Academy of Sciences
Jay I. Kislak Reference Collection (Library of Congress)
Publisher: The Academy,
The Academy
Place of Publication: New York N.Y
Publication Date: 1926
Frequency: completely irregular
Subject: Scientific expeditions -- Periodicals   ( lcsh )
Natural history -- Periodicals -- Puerto Rico   ( lcsh )
Natural history -- Periodicals -- Virgin Islands of the United States   ( lcsh )
Natuurlijke historie   ( gtt )
Geologie   ( gtt )
Expedities   ( gtt )
Genre: bibliography   ( marcgt )
Spatial Coverage: Puerto Rico
United States Virgin Islands
Summary: Includes bibliographies.
Ownership: Provenance: Gift of Jay I. Kislak Foundation.
Statement of Responsibility: New York Academy of Sciences.
Dates or Sequential Designation: Vol. 1, pt. 1-
Dates or Sequential Designation: Ceased with vol. XIX, pt. 1.
General Note: Latest issue consulted: Vol. 18, pt. 4 (1952).
General Note: Kislak Ref. Collection: Vol. 18, pt. 2 (1941)-pt. 4 (1952).
 Record Information
Bibliographic ID: UF00091487
Volume ID: VID00009
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 01760019
lccn - 2002209050

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
        Page i
        Page ii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 96-i
        Page 96-ii
        Page 96-iii
        Page 96-iv
        Page 96-v
        Page 96-vi
        Page 96-vii
        Page 96-viii
        Page 96-ix
        Page 96-x
        Page 96-xi
        Page 96-xii
        Page 96-xiii
        Page 96-xiv
        Page 96-xv
        Page 96-xvi
        Page 96-xvii
        Page 96-xviii
        Page 96-xix
        Page 96-xx
        Page 96-xxi
        Page 96-xxii
        Page 96-xxiii
        Page 96-xxiv
        Page 96-xxv
        Page 96-xxvi
        Page 96-xxvii
        Page 96-xxviii
        Page 96-xxix
        Page 96-xxx
        Page 96-xxxi
        Page 96-xxxii
        Page 96-xxxiii
        Page 96-xxxiv
        Page 96-xxxv
        Page 96-xxxvi
        Page 96-xxxvii
        Page 96-xxxviii
        Page 96-xxxix
        Page 96-xl
    Part 2: Plant ecology of Porto Rico
        Page 97
        Page 97-i
        Page 97-ii
        Page 97-iii
        Page 97-iv
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
Full Text




Porto Rico and the Virgin Islands

Plant Ecology of Porto Rico
H. A. Gleason and Mel. T. Cook

[Issued February 22, 1927]

5-o ,




Porto Rico and the Virgin Islands

Plant Ecology of Porto Rico
H. A. Gleason and Mel. T. Cook

[Issued February 22, 1927]

In trod auction .. .................... ................................. 3
Geography, geology, and physiography ............................ . ... 8
General principles of the development and structure of vegetation. ........ 12
Development of the flora of Porto Rico ......................... . 12
P lant m igration ........................................... ..... 15
The environment ......................... ...................... 20
The plant association............................... ............ 27
The vegetation of the northern coastal plain ............................ 34
G general .................... .................... ............ 34
The mesarch successional series of the limestone hills ................. 36
The mesophytic forest ........................................ 38
The second-growth thickets ............................... . 39
The red clay-loam plains ................................. . 43
The forests of the lowland white sands ......................... 44
Second-growth thickets of the white sands ...................... 46
Reversionary successions on the San Juan formation............. 48
Sum m ary .................................... ............ 49
The xerarch series of sand beaches and sand dunes................... 49
Beaches and beach thickets ............................... . 49
The beach vegetation .................................... 50
The beach thickets ...................................... 50
Sand dune vegetation ........................................ 52
The beach vegetation .................................... 57
The Coccolobis uvifera dunes ........................... . 58
The lee-slope thickets .................................... 60
T he dune forests ........................................ 61
Secondary successions on the dunes .................. ..... 62
The vegetation of the San Juan consolidated dunes......... 65
Disintegration of the San Juan dunes ...................... 67
Sum m ary ..................... ............. .......... 70
The vegetation of Icacos Cay ............................. . 71
The hydrarch series of lagoons and estuaries ...................... 74
The rivers and playa lands ............................... .. 74
Aquatic vegetation of fresh-water lagoons ....................... 76
The mud-bank vegetation .................................... 77
The sand-bank vegetation .................................... 78
T he beach thickets .......................................... 78
Sum m ary .................................... ............ 79
The halarch series of coastal swamps ........................... . 79
The mangrove association .................................... 80
The Acrostichum association .............................. . 84
The Pterocarpus officinalis association .......................... 85
T he clim ax forest ................................ .......... 89
Intermediate vegetation at Fajardo Playa ...................... 89
Vegetation of the Caflo Tiburones ........................... 90
Sum m ary ........................... ........... ........... 96


The vegetation of the central mountain region ........................ 97
General ....................................... ................ 97
Mountain forests at low elevation ................... ........... 104
The forests of higher altitudes.............. ..................... 108
The moist tropical forest . . . ....... .................. 111
The rain forest.............. ......................... 118
The Sierra palm forest . . . .......... ........................ .. 126
The mossy forest.................. ....................... 129
Sum m ary .......................... ... ....... ............... 137
The vegetation of the southern coastal plain and adjacent foothills ........ 139
General ................. ................... ............ 139
The halarch series of coastal swamps and salt flats.................. 141
The mangrove association .......... . . ................. .. 143
The Batis-Sesuvium association ..... . ...... ............... 144
The salt flats.............. .......................... 146
The climax forest........... ......................... 147
Summary ................. ........................... 150
The hydrarch series of lagoons and marshes ...................... 150
The xerarch series .............. ........................ 155
Beaches and coastal thickets............................ 155
The coastal hills.............. ....................... 157
G general ........................ ....................... 157
The xerophytic forest of the Ponce limestone............... 158
The vegetation of the shale hills ....................... 162
The vegetation of the serpentine hills ..................... 165
The vegetation at Cape Mala Pascua..................... 166
General relations and successions ......................... 168
The semi-mesophytic vegetation of the San German limestone 169
The vegetation of the lower mountain slopes ....................... 170
Summary................. .............. ................. 173



The field work upon which the following description of Porto
Rican vegetation is based extended primarily from January 18 to
April 30, 1926, during which period the time of the joint authors
was continuously devoted to it. The general organization of the
survey and some preliminary study were accomplished by the
junior author previously, and both authors have necessarily given
much of their energies since then to the study of material and the
preparation of manuscript.
The opportunity for the survey arose through the hearty co-
operation of the Department of Agriculture and Labor of Porto
Rico, the Insular Experiment Station of Porto Rico, and the New
York Botanical Garden, and has made it possible for the authors
to gratify a long-cherished ambition to prosecute ecological work
in the tropics. They hereby record their appreciation of the
effective interest of Hon. Carlos E. Chard6n, Commissioner of
Agriculture and Labor, of Mr. Francisco Lopez Domfnguez, Direc-
tor of the Insular Experiment Station, and of Dr. N. L. Britton,
Director-in-Chief of the New York Botanical Garden.
The progress of the work has also been facilitated by the assist-
ance and cooperation of Mr. Augusto P. Alvarez, Dr. Jos6 A.
Antongiorgi, Dr. Jaime Bagu6, Hon. Carlos Bahr, Mr. Charles
Z. Bates, Mr. Jorge Bird, Mr. Edmundo Col6n, Mr. Rafael Colo-
rado, Mr. Plhcido Feli6, Dr. Arthur Hollick, Mr. William P.
Kramer, Miss Clara Livingston, Mr. Mariano Mari, Mr. Juan
Masini, Dr. William R. Maxon, Mr. Jos6 I. Otero, Dr. B. E.
Quick, Mr. Julio C6sar Ramfrez, Mr. Virgilio Ramos, Mr. R. A.
Toro, Mr. R. A. Veve, and Mr. Percy Wilson, to each of whom
the authors express their thanks, while regretting that they can
not enumerate the scores of Porto Ricans whose uniform courtesy
and efficient service made the field work pleasant.
Since the work required us to examine all the principal types of
vegetation over a region of some 3400 square miles in a period of


less than four months, it is obvious that no studies of a statistical
or experimental method could be carried on. The method used
was purely that of a field survey. Various areas known to be
occupied by the different plant associations in relatively natural
condition were visited in turn. In each the major superficial
conditions of the environment were noted, the dominant and the
more important secondary species were listed, their inter-relations
described, and the successional trends determined as far as possible
by observation alone.
The structure and appearance of any plant association depend
primarily on the species of plants represented and on the relative
number of individuals of each. The species of nearly every associa-
tion exhibit a variety of vegetational forms, and dominance of one
form, caused by a preponderance of individuals in which it is
developed, determines the general nature of the vegetation. Most
associations of Porto Rico, as also of other parts of the tropics,
contain a large number of species, many of which have the same
vegetational form and accordingly rank as co-dominants. While
any group of them would produce the same general ecological result,
it is nevertheless important to distinguish those which are most
abundant, or which exert the most controlling effect on the environ-
ment, since the abundant species are the ones best adapted to
the prevailing physical conditions. In the Typha-Mariscus as-
sociation, where two species alone form the bulk of the vegetation,
their recognition as dominant species is simple. In a mountain
forest, on the other hand, composed of possibly a hundred arbores-
cent species, a determination of the most abundant and therefore
the most important can not always be accurately made by obser-
vation alone, but requires the application of established statistical
methods, and these we have not had time to use. We have no
doubt, therefore, that more detailed study will reveal cases where
our lists of important species will need change, either because we
omitted some through incomplete field study, or because we in-
cluded some of the less important forms.
Our discussion of the environment has been taken from the
publications of the United States Weather Bureau as to climate,
from the several reports of the Scientific Survey of Porto Rico
and the Virgin Islands as to geology, and from direct superficial
observation of the rocks and soils in the field. In the absence of
all experimental studies of environmental influence, our statements


must be regarded as purely descriptive. While the fact is obvious
that a certain association grows in a certain type of environment,
it may not be true that the association is determined by the condi-
tions which we have mentioned. The fundamental cause of the
association, through which its species have been selected from the
hundreds of possible plant immigrants, may lie in an entirely
different set of factors, possibly active now or possibly operative
during some past period.
In determining the probable trend of succession, we have fol-
lowed well-established principles of ecology and have also been
guided by the known processes of physiography and by the effective
activities of vegetation, so far as we could observe them in the
field, such as soil deposition, humus formation, and control of
The accuracy and completeness of all our work has been lessened
by the effect of man on the plant life. The dense rural population
of Porto Rico, through its constant use of almost all available
land for dwellings, agriculture, pasture, lumber, and fuel, and
through the secondary effect of fire, of drainage, and of irrigation,
has destroyed some types of vegetation completely and, with a
few exceptions, has modified all the rest to a greater or less degree.
Names of plants in an ecological publication are of the greatest
and most fundamental importance, and the value of the work
depends largely upon the accuracy with which the plants are
identified. In this respect the present work is singularly fortunate.
Specimens of almost every species mentioned in our text were
actually collected, preserved, and later identified; the ferns by
William R. Maxon, of the National Herbarium at Washington, the
grasses by A. S. Hitchcock, of the Bureau of Plant Industry at
Washington, and the remainder by N. L. Britton and Percy
Wilson, authors of the Descriptive Flora of Porto Rico (Scientific
Survey of Porto Rico and the Virgin Islands, Vol. 5; Vol. 6, parts
1-3, to which reference is made for descriptions and natural dis-
tribution of the species), at the New York Botanical Garden.
All names of plants follow the usage of the Descriptive Flora.
Our field work extended to all parts of the island and included
detailed study of every type of vegetation known to exist in the
region. In the selection of areas for study we benefited greatly
by the advice of Dr. N. L. Britton and the personal guidance of
Mr. William P. Kramer and Charles Z. Bates. More than half


of the time was devoted to the coastal plains and shores, where
vegetation is best preserved in a natural condition, rather than to
the central mountainous region of the island, where most of the
land is in cultivation or under a poor type of second-growth forest.
Although Porto Rico has received attention from botanists for
many years, work of an ecological nature has been very fragmentary.
The establishment of a national forest reserve and of insular forests
has led to a few descriptive articles, some of which are of great
value. Scattered ecological notes, mostly referring to the habits
or behavior of a single species, occur through a large number of
articles. Urban has discussed the phytogeographical affinities of
the flora. He also published a flora of the island, which has been
much extended and emended through the numerous expeditions
from the New York Botanical Garden, culminating in the Descrip-
tive Flora of Porto Rico, by Britton and Wilson, now recently com-
pleted for the higher plants. This work and the several geological
surveys, now covering nearly the whole surface of the island, afford
the principal bibliographic foundation upon which the present
ecological survey rests. A narrow strip of land extending from
Arecibo on the north coast to Ponce on the south has also been
surveyed by the Bureau of Soils. The following list makes no pre-
tensions to completeness, but merely includes a series of publications
which have some ecological value, or have been used by us.

1. BERKEY, CHARLES P. Introduction to the geology of Porto Rico. Sci.
Surv. Porto Rico 1: 11-29. 1919.
% 2. BRITTON, N. L. Botanical exploration in Porto Rico and islands adjacent.
Jour. N. Y. Bot. Gard. 15: 95-103. 1914.
1 3. Botanical exploration of Porto Rico and the Virgin Islands. Ibid.
24: 93-99. 1923.
*4. Botanical investigations in Porto Rico. Ibid. 23: 49-59. 1922.
5. Botany and horticulture of Porto Rico and the Virgin Island;.
Ibid. 26: 97-102. 1925.
* 6. -- Cactus studies in the West Indies. Ibid. 14: 99-109. 1913.
,7. Descriptive flora of Porto Rico and the Virgin Islands. Ibid. 25:
129-135. 1924.
* 8. Further botanical exploration of Porto Rico. Ibid. 16: 103-112.
* 9. La vegetaci6n de Cayo Icacos. Revista de Agricultura de Puerto
Rico 12: 91-96. 1924.
* 10. -- Recent botanical explorations in Porto Rico. Jour. N. Y. Bot.
Gard. 7: 125-139. 1906.


S11. BRITTON, N. L. and PERCY WILSON. Descriptive flora-Spermatophyta.
Sci. Surv. Porto Rico 5: 1-626. 1923, 1924; 6: 1-371. 1925, 1926.
0 12. BRITTON, N. L. and G. N. WOLCOTT. Porto Rico and the Virgin Islands.
/Naturalist's guide to Americas 700-705. 1926.
/ e 13. CHARD6N, CARLOS E. Flora. El libro de Puerto Rico 36-53. 1923.
14. COOK, 0. F. A synopsis of the palms of Porto Rico. Bull. Torrey Club
28: 525-569. 1901.
e 15. FETTKE, CHARLES R. The geology of the Humacao district, Porto
Rico. Sci. Surv. Porto Rico 2: 117-197. 1924.
16. GIFFORD, JOHN C. The Luquillo Forest Reserve, Porto Rico. U. S. D.
A. Bur. Forestry Bull. 54: 1-52. 1905.
iburvey from Arecibo to Ponce, P. R. Field Oper. Bur. Soils 1902: 793-839.
18. HODGE, E. T. Geology of the Coamo-Guayama district. Sci. Surv.
Porto Rico 1: 111-228. 1920.
19. (HOWE, M. A.) Report on a trip to Porto Rico by Dr. M. A. Howe,
assistant curator. Jour. N. Y. Bot. Gard. 4: 171-176. 1903.
420. HOWE, M. A. Report on a visit to Porto Rico for collecting marine algae.
Jour. N. Y. Bot. Gard. 16: 219-225. 1915.
w 21. HUBBARD, BELA. The geology of the Lares district, Porto Rico. Sci.
Surv. Porto Rico 2: 1-115. 1923.
22. LOBECK, A. K. The physiography of Porto Rico. Ibid. 1: 301-379.
4 23. MAXON, W. R. Descriptive flora-Pteridophyta. Ibid. 6: 373-521.
24. MITCHELL, G. J. Geology of the Ponce district. Ibid. 1: 229-300. 1922.
4 25. MURPHY, Louis S. Forests of Porto Rico; past, present, and future, and
their physical and economic environment. U. S. D. A. Forest Service Bull.
354: 1-99. 1916.
26. SEMMES, DOUGLAS R. The geology of the San Juan district, Porto Rico.
Sci. Surv. Porto Rico 1: 33-110. 1919.
-27. SHAFER, J. A. Collecting in the mountain region of eastern Porto Rico.
Jur. N. Y. Bot. Gard. 16: 33-35. 1915.
28. STEVENS, F. L. Collecting plants in Porto Rico. Ibid. 17: 82-85. 1916.
*29. (UNDERWOOD, L. M.) Report of Professor L. M. Underwood on a trip
to Porto Rico. Jour. N. Y. Bot. Gard. 2: 166-173. 1901.
30. URBAN, I. Zur Pflanzengeographie Portoricos. Symb. Antill. 4: 675-
689. 1911.

Porto Rico is the smallest and easternmost of the four islands
known as the Greater Antilles. It is located between 170 54' and
18 31' north latitude and between 650 15' and 67 15' west longi-
tude. It is approximately 182 kilometers (113 miles) long and 66
kilometers (41 miles) in maximum width, with an area of about
8900 square kilometers (3425 sq. miles).
The Atlantic Ocean a short distance north of Porto Rico is
approximately 7500 meters (25,000 feet) deep, and depths of about
4500 meters (15,000 feet) are reached in the Caribbean Sea to
the south. The channels between Porto Rico and the adjacent is-
lands to the east and west, on the other hand, are only about 200
meters (600 ft.) deep. The island may accordingly be described
as a portion of a chain of mountains, other portions of which are
under water or constitute other West Indian islands. Porto Rico
is itself largely mountainous. A mountain range covering an
average of two-thirds of the width of the island extends its full
length from Panduras Ridge south of Yabucoa to Mayagtiez and
Rincon. Along this range the lowest passes are 600 meters (2000
ft.) above sea level, amd several of the highways rise to even greater
altitudes in crossing. The highest summits are near the center
of the range, where several peaks rise above 1200 meters (4000
ft.) and Cerro de la Punta, the highest point on Porto Rico, reaches
an altitude reported to be 4429 feet (1350 m.). Another moun-
tainous area, the Luquillo Mountains, occupies the northeastern
part of the island. It rises to a height of about 1060 meters (3500
ft.), and is separated from the main range to the south by a low
divide, followed by the main highways between Caguas and
Humacao, and little if any more than 150 meters (500 ft.) in
The crest of the chief mountain range, from Yabucoa to Maya-
giiez, lies well south of the center of the island. The rivers running
north into the Atlantic Ocean are accordingly much longer than
those running south and carry much larger volumes of water,
since the rainfall on the northern slope is twice as heavy as on the
southern side.
The studies on the geology of Porto Rico, published in earlier


numbers of this series and covering about two-thirds of the island,
show that the island consists of three well-defined parts. The
central portion, extending from east to west for the full length of
the island and covering about two-thirds of its width from north
to south, is the oldest. It is primarily of volcanic origin and marks
the site of extensive volcanic activity during the Cretaceous
period by which the land was covered deeply with tuffs, ashes, and
other volcanic ejecta and various extrusive rocks were brought to
the surface, Some of these ejecta fell on the water of the adjacent
Caribbean Sea and formed extensive deposits of shale, which have
since been elevated above the water and appear as the shale hills
of the southwestern part of the island. Similar deposits may have
been made under the waters of the Atlantic Ocean on the north
coast: if so, they do not at present appear at the surface. In
addition to rocks of volcanic origin, the central part of the island
exhibits local deposits of limestone. North of Juana Diaz these
ancient limestones cover several square miles and are a conspicuous
feature of the land.
During the early part of the succeeding Tertiary period, much
of the island was submerged beneath the ocean, so that its area
was reduced to about two-thirds of its present size. At this
time deposits of limestone were laid down along the shores and
attained a great thickness. Another period of elevation followed,
and much of the limestone was lifted above sea level. This eleva-
tion affected the north side of Porto Rico more than the south, so
that all of the island north of a line extending approximately
from Loiza to Aguadilla and passing near Rio Piedras, Corozal,
Ciales, Lares, and San Sebastian has a foundation of limestone,
while on the south shore the limestone area is confined to a narrow
strip extending from near Juana Diaz to Boquer6n, its northern
boundary passing near Yauco and Pefiuelas and just north of Ponce.
Since that time there have been other elevations and subsidences
of less importance. A subsidence of local importance has sub-
merged portions of the north shore in comparatively recent times
and there is some evidence that it is perhaps not yet entirely
Such geological events as Porto Rico has witnessed are of com-
paratively little direct importance in determining the nature of
the present vegetation, except in so far as they have from time to
time restricted or favored the immigration of plants from outside


sources, but through the elevation of the land they have affected
the climate of the island, and through the different kinds of rocks
deposited during their course they have produced different types
of soil and physiography. It will be shown that the three chief
vegetational regions of Porto Rico are correlated with soil or
climate and therefore owe their differentiation primarily to geologi-
cal history.
The land surface of Porto Rico has always been subject to
erosion by the waters of its numerous streams, and the present
configuration of the land is chiefly due to this agency. The central
volcanic region of the island, being the oldest and being composed
largely of volcanic ejecta, has been especially heavily eroded.
The whole region is deeply scarred by innumerable valleys and
ravines, with narrow bottoms and steep sides and separated by
narrow, knife-like ridges. Level plateaus do not exist, although
traces of ancient peneplains may still be discovered. Alluvial lands
along the streams or level flood-plains have scarcely been formed,
although they exist to a small extent along the lower course of some
of the larger rivers which have reached base level near the margin
of the mountain region, and other small areas of alluvial land occur
along a few smaller streams where local conditions have favored its
Notwithstanding the long period of erosion under heavy rain-
fall to which the central part of the island has been subjected,
there is surprisingly little wash or erosion at the present time.
Agriculture is carried on successfully on very steep slopes with no
serious loss by erosion. This is especially conspicuous between
Cayey and Aibonito, where tobacco is grown on hillsides lying at
an angle approximating forty-five degrees. Berkey attributes the
remarkable stability of the soil to three factors: "One is the cling-
ing character of some of the vegetation which tends to bind the soil
together; another is the small range of temperature variations
which reduces disintegration or disruptive tendencies to a minimum;
and still another is the low content of inert or refractory materials,
such as quartz, in the rocks whose destruction has furnished the
soil; all of these factors favor the making of specially tenacious soil."
Both surface and subterranean erosion have been active in the
limestone district along the north shore. In the drier western
end, from Aguadilla east to Hatillo and Lares, dissection has been
less active, leaving a rolling, plateau-like surface. From Lares


east, the limestone has been eroded into a series of isolated hills
with steep sides and narrow or conical tops and separated by areas
of relatively level soil. These hills, variously described as "pepi-
nos," "mogotes," or "haystack hills," are a striking feature of the
topography between Bayamon and Lares. North of the series of
hills the limestone has been almost wholly removed and the land
consists chiefly of alluvial soil deposited by the numerous rivers
which flow north to the Atlantic Ocean. Immediately along the
shore there are also limited areas of sand dunes of both ancient
and modern origin.
There are extensive areas of level land along the south shore
also, principally between Ponce and Guayama, but they have
apparently been formed chiefly as littoral deposits during a period
of comparatively recent elevation. Porto Rico also includes
within its shores small and local areas of fresh water lakes and
marshes, salt water swamps, salt flats, and rock cliffs. All of these
types of topography will be discussed further, so far as they affect
plant life, in connection with the vegetation which occupies them.
Porto Rico lies in the path of the north-east trade winds, which
blow with great regularity, Their influence is greatest along the
north coast, where they serve to keep the extremes of temperature
within a very moderate range. They also account for the heavy
rainfall north of the divide and the low rainfall along the south
coast. Along the north coast they cause the construction of
sand dunes and the development of one-sided trees and often
interfere with agriculture in the immediate vicinity of the ocean.

The oldest rocks now contributing to the surface of Porto Rico
are of Cretaceous age and partly of volcanic origin. Fossils of
this period from other parts of the world show that during the
Cretaceous period there was already in existence a rich flora of seed-
plants and that they had attained a wide distribution over the
then land surface of the earth. There is also reason to believe that
by the close of the Cretaceous the neotropic and the paleotropic
floras were also differentiated. There is no cause for doubt,
therefore, that during this long period of time there was ample
opportunity for a phanerogamic flora to reach Porto Rico by
migration and to develop upon the island a forest vegetation.
There is no reason to suppose that the volcanic activities which
characterized this period in Porto Rico were so continuous either
in time or in space, that they prevented this immigration and
colonization, since experience with other volcanic deposits of
modern time has shown that only a very short period is necessary
for the establishment of vegetation after a volcanic eruption.
So far, we have no evidence from Cretaceous fossils concerning the
nature of this early flora.
Some minor volcanic action probably occurred during the suc-
ceeding Tertiary time, and the whole period since the close of the
Cretaceous has been marked by various oscillations in level.
Geological evidence shows that at no time has the whole island been
submerged since the Cretaceous, and the extent of volcanic action
seems to have been so limited that there is every probability that
the phanerogamic vegetation of Porto Rico has been continuous
to the present time.
In general, there are but two possible sources of the Porto
Rican flora, by immigration from another region and by evolution
within the island. It is not within the province of this survey
to discuss either of these sources in detail, but certain considera-
tions are necessary to an understanding of the present vegetation.


Immigration of plants into Porto Rico is entirely independent of
the present relation of the island to other lands in the West Indies.
If at some period in its past history, Porto Rico was physically
connected with other islands to the west or east, immigration
could take place by a slow spread of species overland. If the
island has been disconnected from other lands since its earliest
time, immigration has been possible only by a considerable leap
over the intervening waters. While comparatively few plants
have devices for migration by which they may regularly or fre-
quently cross a water barrier as wide as those now separating
Porto Rico from the other islands to the west and east, nevertheless
the time has been so great that even occasional and accidental
migrations can account for the present flora. If the submergence
or elevation of the islands has widened or narrowed these water
barriers from time to time, the immigration of plants has been
retarded or facilitated, but the entrance of plants into Porto
Rico must have been continuous from the Cretaceous period to
the present time. There is every reason to believe that it is as
rapid now as at any previous period, and that the future will see
still further additions to the native flora.
The most rapid accession of additional species by immigration
is of entirely modern development and dates back only to the
arrival of man in the West Indies. The Caribs are known to
have made frequent passages from one island to another, and they
were undoubtedly the means of introducing, either voluntarily
or involuntarily, numerous species. These include plants used for
food, shelter, utensils, or personal adornment, which were brought
in purposefully, as well as many others which followed the Caribs
in some accidental manner. The arrival of the European four
centuries ago, and the consequent establishment of frequent
connection by ship with other and more distant lands have again
served to aid the immigration of scores of species. Some of these
newer arrivals have also been brought here intentionally, such as
the bucare, Erythrina Poeppigiana (Walp.) 0. F. Cook, but many
others have merely followed man and at the present time constitute
the bulk of the common weeds of the island. Most of these have
never spread beyond the range of human activities on the island,
and are closely restricted to fields, roadsides, and waste places,
where the native vegetation has been destroyed or greatly altered.
The general result of this modern immigration has been to give


Porto Rico many cosmotropical species, so that the weed vegeta-
tion is now closely similar to that of other West Indian islands.
It may also be noted that if immigration has brought species
into Porto Rico, so has the island also served as a source of emigra-
tion to other places. It is entirely possible that many species of
Santo Domingo and the Virgin Islands have reached them by
emigration from Porto Rico. Such an emigration, of course, does
not imply that the the species became extinct in Porto Rico as a
For the vast majority of plants, botanists are still in ignorance
as to the details of time and place of the origin of species. Yet
paleobotanical evidence makes it clear that such an evolution
has been very extensive. The genera and families of plants now
preserved as fossils, and dating back to the Cretaceous or Tertiary
periods, are usually either identical with or closely similar to
existing genera and families, but the species are for the most part
distinct. It is accordingly obvious that evolution since the first
appearance of a Porto Rican flora has been so wide-reaching in
its extent that most of the early species have become extinct and
have been replaced by related species of more recent origin. So
far our knowledge of fossil Porto Rican plants does not enable us
to speak with authority on conditions in this island, but there
is every reason to believe that a similar condition has obtained
here, and that the species composing the flora of the island have
changed almost completely since the early Tertiary time.
Again, it is not within the province of this survey to discuss
any details of evolution among plants, but it must be noted that
there is no reason to suppose that Porto Rico has been either more
or less favored, in comparison with other lands, as a locus for
specific differentiation, except as it offers greater or less variation
in its environment.
If the various possibilities of migration and evolution are col-
lated, it will be seen that the species of Porto Rico may be divided
into several different categories.
A. Species common to Porto Rico and other parts of the world.
1. Originating elsewhere by evolution and reaching Porto Rico by immigra-
2. Originating in Porto Rico by evolution and reaching other lands by
3. Originating both in Porto Rico and in other lands by evolution from
a common ancestral stock.


B. Species endemic to Porto Rico.
4. Originating elsewhere by evolution, reaching Porto Rico by immigra-
tion, and becoming extinct in the original habitats.
5. Originating in Porto Rico and not at present colonized in other lands
by emigration.
With our present knowledge of the vegetation and geological
history of the West Indies, it is not possible to assign most plants
of Porto Rico to their proper place in this arrangement, but care-
ful study and correlation of morphological characters, geographical
distribution, and ecological behavior of the species will in the
future do much to unfold the history of Porto Rican vegetation.

Whether established in Porto Rico by immigration from a
foreign source or by evolution, each species has the whole extent
of the island immediately available for migration and further
colonization. We must distinguish carefully between the two
processes, the former referring merely to the movement of dis-
seminules from the place of origin, that is, the location of the parent
plant, and the latter to the successful establishment of these
disseminules in a new area, their growth into mature individuals,
and their continued reproduction through successive generations.
The capacity of any species for migration depends on the struc-
ture of its disseminules and their ability to utilize the various
sources of energy necessary for their movement. In general, the
capacity of a species for migration has been greatly underestimated.
It is, of course, a matter of common knowledge that certain species,
especially weeds utilizing the agencies of commerce, are able to
travel rapidly, and consequently cover large territories in a sur-
prisingly short time. This is quite obvious along the numerous
highways of Porto Rico. It is equally well known that other
species, especially those with large and heavy fruits, have apparently
no method of rapid migration whatever, so that their disseminules
normally come to rest within a short distance of the parent plant.
Between these two extremes lie the great majority of species,
which have only moderately efficient means of migration. As an
example of the first category, the introduced Asiatic weed Ver-
nonia cinerea (L.) Less. may be cited. Its seeds are produced within
a single year and are easily blown to considerable distances by
the wind. On the other hand, the large fruits of the jacana,


Lucuma multiflora A. DC., fall to the ground beneath the parent
tree and require several years for their growth into a second genera-
tion of mature seed-bearing plants. If no other factors were taken
into consideration, it would appear that Vernonia cinerea may have
reached the island a short time ago, which is of course true, and
have attained its present broad distribution quickly and easily,
and that the jacana, to have attained its equally wide distribution
must have been a plant of early appearance in Porto Rico. On
the basis of such reasoning Willis formulated the law of Age and
Area, which states that in general the area of a species is an
index to its age.
But the effect of accidental long migrations must be taken into
consideration. Every species of plant occasionally profits by such
accidents. Storm winds pick up seeds and carry them long dis-
tances beyond their usual range of migration. Fruit-eating birds
and mammals carry them away and frequently lose them before
they are devoured. Torrents of water, following heavy rains,
wash seeds to great distances. These and other similar forces of
occasional occurrence all combine to give every species, at some
time or other, a range of migration far in excess of the normal.
It is true that the great majority of individuals of any species
develop as a result of a slow normal migration, but the extension
of the range of the species depends more upon the longer accidental
movements. It is probable that the average fruit of jacana comes
to rest on the ground not more than twenty meters from the
parent, and that at least ten years are required for its growth
into a seed-bearing tree, so that its ordinary movement may be
two meters per year or 500 years per kilometer. At this rate
80,000 years would be required for the species to extend its range
from one end of the island to another. But a single fruit, carried
by some fruit-eating animal, might easily accomplish in a single
generation a full kilometer and a series of such events reduce the
total time to as low as 1600 years. No actual statistics are avail-
able on this or similar points, but the known rapidity of migration
of many species makes it apparent that the far-reaching effect
of accidental migrations must not be overlooked.
It is also apparent that species immigrating to Porto Rico may
have done so not once only, but many times, and established them-
selves at various points on the island, thereby shortening the
necessary time for their complete occupation of the territory, and


that species arising by evolution in Porto Rico may have appeared
not as single individuals but as groups of individuals with a con-
siderable range of initial distribution.
It is, on the other hand, perfectly conceivable that some Porto
Rican species may have reached the island or have evolved so
recently that they have not yet had time to migrate beyond their
region of first appearance.
Immigration and evolution have probably gone on continuously
since the early Tertiary period at least, which was certainly more
than a million years ago, and have probably been no more active
at one period of this long time than another. Since the minimum
time for complete migration may be as low as a few centuries, it
becomes apparent that only a very few plants have not yet had
sufficient time for their full migration over the island. Even the
exotic weeds which have entered Porto Rico during the last four
centuries seem to have in general ample time, since their migration
is regularly rapid enough to compensate for their recent intro-
duction. We may accordingly safely conclude that the vast
majority of species have lived in Porto Rico long enough to have
enabled them to migrate throughout the island.
Field studies, however, show at once that there are many native
species in Porto Rico which have not attained this complete
distribution and which are still limited to particular parts of the
island. The Luquillo Mountains are noted for the large number
of local species not known elsewhere; even the rocky summit of El
Yunque, barely half an acre in extent, has at least three endemic
species. The reason for this is not lack of migration, but the
failure of the seed to establish itself successfully in its new site,
or else the existence of barriers between suitable sites sufficiently
wide to prevent the migrating species from crossing them. The
former condition is of chief importance in an island as small as
Porto Rico, but the effect of the latter is also discoverable.
Every species of plant requires a certain complex of conditions
for the germination of its seeds and their subsequent successful
growth. These conditions include every feature of the environ-
ment which affects the physiological processes of the plant, such
as composition of the soil, soil water, atmospheric humidity, light,
temperature, air movement, as well as soil organisms and parasitic
diseases. Flowering plants are in general not exacting in their
requirements but withstand a considerable amount of variation


in each of these. But if any one of these factors or the combined
effect of two or more of them exceeds the tolerance of the plant,
the seed does not germinate, or the young plant dies before reach-
ing maturity. As a result, each type of environment is character-
ized by those particular species which can succeed under those
available conditions, and all other species are excluded, no matter
how often their seeds may immigrate.
It is never possible to determine from field observation alone
whether the difference in flora between two widely separated
habitats of apparently similar environment is due to minor and
obscure environmental differences or to the failure of migration,
due to insuperable barriers. As an illustration, the floras of El
Yunque peak at the eastern end of the island and of Cerro de la
Punta in the center of the island may be cited. Both of these
are high enough to be relatively cool, to condense a heavy rainfall,
and to be well exposed to wind, and both are occupied by essentially
similar vegetation. It is obvious, however, that the latter has the
less wind, the lower rainfall, and the drier atmosphere. Of the
four species of shrubs or low trees which characterize the vegetation
of the summit of El Yunque, two are absent from Cerro de la
Punta. Their absence may be due to these environmental dif-
ferences, although the differences are not sufficient to exclude the
other two species. Careful experimental studies would be neces-
sary to determine this, although we are prone to infer a different
cause, since the floras of the two regions are essentially similar.
Between El Yunque and Cerro de la Punta, with the adjacent
high peaks, is a gap of some 70 km., within which conditions
suitable for the growth of these species of high altitudes do not
exist. It is not possible for any of them to migrate from one region
to the other by slow stages across the intervening lowlands: they
must have bridged the gap in some unusual or accidental method
by a single flight. It is entirely possible that the two missing
species have never yet had the advantage of such an unusual
event, and that they may at some future time appear on Cerro de
la Punta. Numerous instances of similar distribution occur in
Porto Rico. Any attempt to explain them must be purely gratui-
tous in our present state of knowledge. To cite barriers as a
reason must depend entirely on inference and to cite environmental
differences needs careful substantiation by detailed environmental
measurements and physiological experiments.


On the other hand, there are some instances easily observed in
the island that leave little or no doubt as to the nature of the in-
hibitory cause which prevents complete distribution of a species.
Ferns of the genus Dicranopteris are common and characteristic
of the central mountain mass of the island. Their microscopic
spores are distributed far and wide by the winds, and young plants
appear promptly on any new and favorable situation. Normally
inhabitants of rocks and cliffs within the forest, they have colonized
immediately the banks exposed in road-making, and there is no
reason to doubt that their spores have every year migrated freely
to the limestone cliffs of the haystack hills immediately north of
the mountain mass. Yet the plants have not developed there,
and the conclusion is obvious that the considerable difference in
the soil of the two regions has been the inhibitory factor.
A similarly characteristic species of the limestone hills of the
north side of the island is the endemic palm, Gaussia attenuata
(Cook) Beccari. Its tall, slender stems appear as scattered individ-
uals toward the summits of many, but not all, of the haystack hills.
Young plants may be discovered on some of these hills where mature
individuals are not present, and the species reappears on the lime-
stone hills of the southwestern part of the island, showing that
it has migratory ability. Its absence from many of the haystacks
is accordingly best explained by the difficulty of migration from one
summit to another, rather than by any environmental difference,
and we may expect that it will appear, from time to time, on any
of the haystack hills as the accidents of migration bring about
its introduction.
The actual origin of the flora of any one habitat is therefore a
very complicated matter, and may again be illustrated by one of
the haystack hills. An average hill is occupied by some 200
species of flowering plants. The great bulk of seeds produced by
this population do not migrate beyond the same hill, and we may
well believe that the vast majority of existing mature plants
were derived from seeds produced on the same hill. But there are
undoubtedly numerous cases every year of seeds migrating in
from adjacent hills. Whether this interchange of species is
sporadic or continuous, frequent or rare, depends on the migratory
ability of the species, but as a result of it, all of the hills have in
time developed an essentially similar flora and vegetation. Many
species, such as the Dicranopteris ferns, certainly migrate to the


hills every year, while others appear at longer intervals, but are
unable to establish themselves because of unsuitable environmental
conditions. There may well be still other species in the island
adapted to the haystack environment which have not yet succeeded
in migrating to them because of too great intervening distances,
but which may appear at some future time.

In the preceding section, emphasis was placed on the fact that,
in most instances at least, no single environmental factor can be
cited as the determining reason for the presence or absence of any
one species, but that all features of the environment act together
as a complex unit to favor the development of some species and to
prevent the growth of others.
The number of different environmental complexes which exist
in Porto Rico is great. The rainfall varies as from five to one;
atmospheric humidity from high, as on the Luquillo mountains, to
low on the hills at Guayanilla; the wind from almost constant to
very little; the soils from calcareous to acidic, from coarse and dry
to fine with a great capacity for water, from purely inorganic
to almost purely organic, from stable to shifting; the soil water
ranges from little in the crevices of vertical cliffs to complete and
constant saturation in swamps. Physiographic processes are here
creating new habitats, there destroying old ones, and here again
modifying others. And with all of these physical, inorganic
conditions, the plant cover itself is constantly at work, changing
the chemical and physical nature of the soil, producing humus,
reducing the wind and light, increasing the atmospheric humidity.
In the older parts of the island these physiographic and vegetational
processes have been in operation for thousands of years, and each
change in the general environmental complex has been marked
by a corresponding change in the vegetation, which in turn served
to alter or to control the environment further. In these places, the
environment has generally reached a condition of relative stability
and consequently has also a stable type of vegetation which will
change but little or not at all with the lapse of years. In other
places, the change in the environment is still proceeding rapidly, and
each effective change is marked by a corresponding change in the
vegetation, which in turn becomes a part of the environment, tend-
ing in some places to retard, or in others to accelerate further change.


In still other places, but limited in their extent, new sites for plant
life are appearing and just starting on their long cycle of vegetative
change and development, perhaps to follow it through to its com-
pletion, or perhaps to be destroyed in the near future.
To each of the various habitats which exist for vegetation in
Porto Rico there is given a foundation of certain environmental
conditions which the plant life itself can not effectively change.
Each receives an amount of heat determined by broad climatic
conditions; each receives rainfall and is exposed to wind as deter-
mined by its location and topography; each receives essentially
the same amount of light from the sun. These unalterable char-
acters, however, are but a small part of the whole environmental
complex, and the larger part is in almost every case subject to
great modification by the presence of plants. The taller species
intercept much of the light and require the smaller species and their
own seedlings to develop in shade. They reduce the penetration
of wind beneath them, thereby tending to increase the atmospheric
humidity and to reduce transpiration. They change the nature
of the soil both chemically and physically through the action of
their roots and through the accumulation of fallen leaves and
twigs. In these and in other ways, the plants immediately be-
come a part of the environment and unite with the purely physical
factors to create the total environmental complex and thereby
to determine the vegetation of every habitat. The influence of
the plant life varies greatly in its importance, approaching a
minimum on the sand dunes along the coast and a maximum in
the mountain forests, but it is always present, and each acre of
Porto Rican land portrays in its present vegetation the result
of this long-continued interaction, extending in some cases back
to the Cretaceous period.
So important are both the physical and the vegetational features
of the environment to the development of the existing vegetation
and so slow has been its genesis, that the destruction of the present
vegetation by artificial processes is in most cases followed by
the appearance of an entirely new type of plant life. An abandoned
field in the mountain districts is soon covered, not with forest,
but with dense thickets of fern. The clearing of vegetation around
a haystack hill is followed by the development of a thicket of
peppers. Only a flooded mangrove swamp, to which very few
species are adapted, regenerates itself at once after cutting. The


details of the underlying causes which control the determination
and the alteration of vegetation are in most cases unknown, and
for Porto Rico, in the present state of our knowledge, can only
be inferred, although our inferences are in many cases well justified
by our observations.
In the following description of the vegetation of Porto Rico
the leading features of the environment are presented in turn for
each type of plant life, but the broader features of temperature
and rainfall are discussed in this place.
The rainfall of Porto Rico is derived chiefly from moisture-
laden winds approaching the island from the north or east. These
condense and precipitate their moisture chiefly on the northern
two-thirds of the island, which lies north of the main divide.
Probably the maximum rainfall, and certainly the most continued
humidity, is received on the slopes of the Luquillo Mountains,
occupying the northeast corner of the island. The amount of
rainfall here is not fully known, but approximates 135 inches (340
cm.) per year. Toward the west on the north side of the island,
there is a gradual but slight diminution in rainfall. While Rio
Blanco, near the Luquillo Mountains, receives slightly over 100
inches (250 cm.), San Juan has an average rainfall of about 65
inches (160 cm.), and.Isabela not far east of the northwest corner,
has only 54 inches (135 cm.). Inland, as far as the divide, the
amount varies, but does not average less than 60 inches (150 cm.),
and in many stations exceeds 80 inches (200 cm.). The western
end of the island, even at low altitudes, also has a similarly heavy
South of the main divide, there is an abrupt and important de-
crease in rainfall, which reaches its minimum over the western half
of the south coast. Throughout this region, the average precipita-
tion is less than 60 inches (150 cm.), in some stations less than 40
inches (100 cm.), and the minimum for observing stations with
records of ten years or more is at Potala, with an average of only
32 inches (80 cm.). When it is considered that, in general, 80
inches is necessary for the development of a true rain-forest in the
tropics, and that a rainfall of less than 40 inches can seldom support
a more luxuriant vegetation than desert scrub, it will be seen that
the rainfall must have a profound influence on the vegetation and
the distribution of species in Porto Rico.
The average annual rainfall of the island is shown on the map


40 40 GO
FIG. 1. Average annual rainfall of Porto Rico in inches. The region of protracted drought along the south shore is
shown by vertical lines; horizontal lines indicate the drought-free region of the Luquillo Mountains.


(Fig. 1), compiled from the records of the United States Weather
Bureau in San Juan. The map also shows the region in which
protracted droughts occur every year.
The transition from mesophytic montane forest, extending from
the north side of the island to a short distance south of the divide,
to a xerophytic scrub is remarkably abrupt, and a distance of two
or three kilometers is sufficient to make a radical change in the
general aspect of the vegetation. This transition zone is so narrow
that the amount of rainfall alone can scarcely be invoked as an
explanation, and the atmospheric humidity may also be of con-
siderable importance. The relation of atmospheric temperature
and humidity to altitude is well known. As air currents rise in
passing over the mountains from north to south, they become
cooler and lose a part of their capacity for holding water vapor,
which is accordingly precipitated as rain. As they descend on the
south side, the increasing pressure raises their temperature and
increases their water-holding capacity, without providing the
necessary water to saturate them. The winds of the south side
are therefore both warm and dry, and this, through its effect on
the transpiration of the plants, certainly acts with the deficient
rainfall to favor a xerophytic vegetation. Unfortunately, no
records are available to show the comparative evaporation of
water from a free surface on the two sides of the island.
Still another feature which is detrimental to a mesophytic
vegetation on the south side is the frequent long periods of drought.
The rainfall is irregular and subject to great variation throughout
the island, but the condition is emphasized on the south side, and
may be illustrated by the reports of the observing station at Potala.
During the nine-year period 1914 to 1922, inclusive, there were 88
periods of drought extending over ten days or more and with an
average length of over 21 days. These periods extended to as
much as 80 days, and even in the best year there was one period
of 28 days, while the average maximum for the nine years was 51
days. The condition is most pronounced during the four months
December to March, inclusive, each of which has from 22 to 27
days of drought, on the average, and reaches its minimum during
September, in which the average drought is only 7 days, the
remainder of the ten-day periods being located in August or October.
It is clear that a normal mesophytic vegetation can scarcely with-
stand such conditions, unless unusual supplies of soil-water are
available to compensate for the deficient rainfall.


Every land presents a variable temperature, in which four sorts
of variability may be traced. Each day shows a variation from a
minimum, usually shortly before sunrise, to a maximum, usually
shortly after noon. There is a variation from day to day in the
mean daily temperature, as warmer days alternate with cooler ones.
There is a variation from season to season, depending in general
on the length of the day and the altitude of the sun above the hori-
zon, and lastly, there is a variation from year to year, as cooler
years alternate with warmer ones. In all of these, Porto Rico
exhibits great stability of temperature, especially in comparison
with a temperate zone and a continental climate. Lying wholly
within the tropical zone, surrounded on all sides by water, and
under the well-nigh constant influence of the easterly trade winds,
all of these fluctuations are greatly reduced. The usual range of
temperature through the course of a single day is in general about
200 F. at the coast, and as much as 300 at some of the inland stations.
The mean daily temperature seldom changes as much as 50 F.
from one day to the next; the difference in average temperatures
between the warmest and the coldest month is only 5-8 F., and
the fluctuation in mean temperature from year to year does not
exceed one degree. It is obvious that such conditions as these
can exert no appreciable effect on the distribution of plant life.
Of far greater importance is the variation in temperature with
the altitude of the land. Meteorological stations are maintained
in the island at elevations ranging from sea level to about 600
meters (2000 feet), and show a deviation in mean monthly tempera-
ture of five to seven degrees below that at sea level. At the
coastal stations, the minimum temperatures occasionally drop
below 600 F., but in the mountain stations temperatures below
500 are frequently reported. Several of the mountains rise above
1200 meters (4000 feet), but no records are available concerning
temperatures on their summits. These peaks are of course colder,
but decrease in temperature above the height of the higher plateaus
is doubtless less, due to the effect of temperature inversions, and
the character of the vegetation on even the highest summit makes
it appear improbable that freezing temperatures ever occur.
Accompanying the gradual reduction in temperature from low to
high altitudes, a variation in the flora may also be discovered.
Numerous species of the coastal region do not ascend much above
sea level, such as the common Bucida Buceras L. of the southern


coast, or Stahlia monosperma (Tul.) Urban of the east and west
ends of the island. At an altitude of about 600 meters (2000 feet)
in the Luquillo mountains, and about 900 meters (3000 feet) in
the central Cordillera, forests of the sierra palm appear, alternating
with mossy forest, and extend to the highest elevations. But the
diminution in temperature is so gradual, and the lower boundary
of the palm forest and ridge forest is so abrupt, that temperature
can not be considered the sole explanation. In general, a species
of plant can grow and reproduce in temperatures both higher and
lower than the extremes to which it is subjected through its natural
range, and it appears as a consequence that altitudinal limits are
fixed by other causes, among which the temperatures may be a
contributory factor. Especially, atmospheric humidity, rainfall
and consequently soil moisture, and wind increase with the altitude,
while periods of drought decrease in length and frequency.
Nor can we assume that even this complex of environmental
conditions always actually favors directly the development of
certain species at high altitudes. It may act instead indirectly
through the inhibition of other species, thereby leaving more space
for the development of those characteristic of the high elevations.
Where a large number of species exist, all about equally adapted
to the environment, a rich vegetation is produced in which large
numbers of species are represented by relatively few individuals
of each. But at higher elevations, where many species are in-
hibited by the united effect of increased humidity, rainfall, and
wind, and the decreased temperature, the remaining species develop
more individuals and thereby produce a characteristic vegetation.
The probability of the effect of this condition, rather than of tem-
perature alone, is illustrated by the lower altitude of the sierra palm
belt on the Luquillo mountains. Their location at the northeast
corner of the island gives them full exposure to the moisture-laden
easterly winds, so that the effect of increased wind, humidity, and
rainfall appears at a lower elevation than in the central Cordillera.
The generally xerophytic nature of the vegetation of the southern
coast of Porto Rico has already been mentioned, and there also,
temperature is a contributing factor. In this part of the island, the
daily maxima are higher than on the north shore, so that plants are
regularly exposed, during their hours of transpiration, not only to
a greater insolation and a drier atmosphere, but also to a higher


In summary, we find that in Porto Rico the great climatic
features of rainfall, atmospheric humidity, wind, and temperature
have a distinct relation to the topography; that all of them exert
an influence on the distribution of species and on the grouping
of species into distinct types of vegetation, but that no one of
them alone can offer an adequate explanation of the nature or
distribution of the plant life of the island. How far each of these
climatic factors contributes to the allocation of the vegetation can
be determined only by the most careful experimentation.

Broadly defined, plant ecology is the branch of botanical science
which deals with the relation between the plant and its environ-
ment, including under the term environment every factor or
condition which affects the life of the plant in its natural surround-
ings. The effect of environmental influence on the plant is made
manifest in many ways, including collectively all its features of
structure and behavior which are not inherited from the previous
generation. These environmental effects may be classified under
three general heads.
Firstly, the environment affects, and to a certain extent controls
and determines, the structure and behavior of each plant individual.
For example, the shape of the leaf on any species is in general
inherited, but the size, the thickness, the amount of chlorophyll,
the number of stomata, and the structure of the vascular bundles
may vary with different amounts of light to which the leaves are
exposed and with different amounts of water supplied to them from
the stem. Not all plants exposed to the same environment present
the same structural features as a result, but in many types of
environment certain structural peculiarities are shared by many
of the plants, so that the vegetation presents a prevailingly uni-
form aspect.
Secondly, no species of plant is able to grow in all, or even in
most, of the available types of environment, and for all species
there is a certain range of environment within which only the plant
can successfully grow and reproduce. As a result, each area of
ground in which a particular complex of environmental conditions
is repeated tends to be occupied by the same group of species.
The individual plants of this area in turn react upon and modify
the environment, not only for themselves, but for the adjacent


plants of other species, so that an interrelation between the plants
is developed. This leads to the formation of definite aggregations
of plants, comparable in many ways to the social or political organi-
zations of man, and termed plant associations. The recognition of
such plant associations, and the study of their structure, their relation
to the environment, their mode of development, and their relation
to other associations in time and in space constitute a broad and
complex field of plant ecology.
Thirdly, the environment affects and controls the distribution of
each species over the world. All individuals of any species have in
general the same environmental demands, and the species is able to
grow only over that part of the world where the necessary conditions
are met. The study of the distribution of species and of the causes
which have produced it is usually known as phytogeography.
Research has shown that many species of plants do not at the
present time occupy the full range to which they are adapted.
Some species have been prevented from extending their migrations
by the existence of barriers which they have so far been unable
to cross, while others are even now in process of migration.
An ecological survey is primarily concerned with the second
phase of ecological study. It deals with the various types of
vegetation found in the region, including their component species,
their ecological structure and general appearance, their correlation
with the environment, and their interrelations in time and space.
It is founded not on the behavior and structure of individual plants,
nor on the broader geographic distribution of species or larger
classificatory groups, but upon the aggregations of diverse species
into areas of definite composition and appearance. It is these
aggregations, generally known as plant associations, which cause
the varying aspects of vegetation over the world.
There are few phenomena of plant life more patent to the general
observer than the fact that the vegetation of any region is neither
uniform throughout nor, in most places, gradually variable from
one place to another. On the contrary, it is usually divided into
numerous types, each of which occupies a small or large area
throughout which it preserves a uniform appearance. These types
are repeated in many places, and the transition from one to another
is frequently so abrupt that it can not fail to attract attention.
No better example need be cited than the swamp vegetation along
the north coast of Porto Rico. There hundreds of acres are covered


by a uniform, homogeneous thicket of mangroves. On the land-
ward side of these mangrove swamps there are equally homogeneous,
but smaller, areas of bracken fern, and at still other places, dense
patches of cat-tails. Although lying side by side, the species of the
mangrove and bracken fern associations do not mingle, and there is
a similar abrupt transition between the fern and the cat-tail
The development and the maintenance of these associations year
after year depends essentially upon two underlying causes, migra-
tion and environment. Into each area now occupied by the pure
stand of mangroves, migration brings every year the seeds of
numerous other species, all in themselves perfectly capable of
growth into mature plants, but the environment in this particular
area effectually prevents the growth of any except the mangroves
and the few other species normally associated with them. If a
new swamp should be excavated, scores of species would at once
migrate into it, and from them the local environment would select
those few which are able to grow there. Thus, if the water was of
proper depth and salinity, the new swamp would soon be colonized
by mangroves, and if nearly or quite fresh by cat-tail and its
associated species.
But the raison d'etre of the association is not entirely so simple,
since the plants themselves tend to control and alter the environ-
ment. Their branches and leaves shade the ground, reduce the
available light, and prevent the growth of species which demand
more light. Their fallen leaves and twigs decay on the ground and
add humus to it. In these and many other ways the plant life
reacts on the environment, so that the choice of immigrating
species is not purely a physical process, but depends also very
largely on the existing plant population.
The general appearance of any association depends on the size,
form, and relative numbers of each species represented. In every
association there are some species, essentially similar in size and
appearance, which far out-bulk the rest and consequently determine
the general aspect of the whole association. In the mangrove as-
sociation, for example, three species of mangrove, all much alike in
size and habit, constitute the mass of the vegetation, while the few
species of smaller plants beneath the trees might all be removed
without altering the general appearance of the association. At the
same time, these species, because of their great number and bulk,


are of the greatest importance in their reaction on the environment.
In almost every association similar important species exist, which
determine the general appearance of the vegetation and control its
environment through their superior size or number, and these are
termed dominant species. The remaining plants, comparatively
small in total bulk, reacting but little on the environment, and
contributing but little to the general aspect of the association, are
known as secondary species.
Environments, as determined by temperature, light, and rainfall,
vary but gradually from place to place, and even changes in the
soil are usually more or less gradual, while the changes in plant life
are usually abrupt. It seems, therefore, that these abrupt transi-
tions in vegetation can not always be explained on a physical basis
alone. Instead, they are to be accounted for by the reaction on and
alteration of the physical environment by the plant life, so that the
gradual variation of certain conditions may be converted into an
abrupt transition by the different reactions of two unlike sets of
species. Thus in the brackish waters of a Porto Rican lagoon there
is no abrupt variation in salinity, but the restriction of free circu-
lation by a dense vegetation and other sorts of environmental con-
trol are sufficient to produce the sharp line between the mangroves
and bracken fern of the more saline waters and the cat-tails nearer
the shore.
Environments of plants are constantly changing. The water of
swamps becomes shallower as the plants deposit muck on the bot-
tom. On some hills rock is disintegrating into soil. On others,
the soil is being washed away by rain, and more bare rock is being
exposed. Every case of environmental change leads to a new
type of environmental selection from the immigrating seeds and
changes the plant population accordingly. As a result, plant
associations have but a limited duration. Sooner or later, the
original species are no longer able to grow in the new conditions, and
their place is taken by a different set of species which now find the
environment which they require for their successful development.
The old association disappears completely and its place is taken by
a new one, which will in turn meet the same fate after a lapse of
years. This process is known as succession.
The duration of a plant association may be short or long, depend-
ing on the speed of environmental change. Good examples of
short-lived associations may be seen along the north coast of


Porto Rico on the sand dunes, many of which are continually in a
state of flux. New types of environment appear from year to year,
and each is immediately occupied by a group of species adapted to
it, only to be destroyed and replaced by others in a very few years
more. But there are other environments which are exceedingly
stable and in which no dynamic processes are now at work which
can change the environment sufficiently to alter the vegetation, or
in which environmental change is so slow that other forces, such
as geological uplift or subsidence or climatic changes, might super-
vene and nullify any predictions which could be made about the
ultimate fate of the present vegetation. For example, the moun-
tains of Porto Rico at high elevations are occupied by forests
different from those of lower levels. The erosion of these moun-
tains by water has been in operation since the Cretaceous period
and has already produced the deep canyons which intersect them.
This erosion is so slow that, before the higher elevations can be
reduced and a new type of vegetation developed on them, further
geological activities, such as renewed uplift or vulcanism, may undo
all the effects of erosion. It is futile to speculate on the future or
the past of such associations, which appear to be almost permanent
and to which the name of climax associations has been given.
In summary it may be repeated that plant associations are the
basic units of vegetation, that they are the result of immigration
and environmental selection, and that their duration is short or
long, depending on the rate of environmental change. The duty
of an ecological survey is to describe these associations, to correlate
them with their environment, and to discuss their past history and
probable future so far as clear evidence exists on which such a
discussion may be based.
A survey of the vegetation of Porto Rico shows at once the
presence of a large number of different plant associations in the
natural vegetation. Some of these are of large extent and were
originally, before their destruction by man, nearly continuous
over wide areas. Others are very small in area and are represented
by scattered and isolated examples. Some show clearly that their
existence is precarious and their duration frequently short, because
of rapid change in the environment which they occupy. Others
offer no evidence that they are subject to change, except as they
have been or may be destroyed by man. Still others show clearly
that they owe their existence to man, through his activities in


cutting, clearing, or burning. For many of the short-lived associa-
tions we can discover, with what we believe to be considerable
accuracy, what their antecedents have been and can predict what
associations will succeed them.
In an area with as many associations as Porto Rico, some system
of classification is necessary. We have accordingly distinguished in
Porto Rico three major groups which are primarily geographical
in nature and which are correlated with certain broad features of
climate and soils. These are first, the vegetation of the northern
coastal plain, occupying a region of limestone or alluvial soils and
characterized by a heavy rainfall; second, the vegetation of the
central mountain region, occupying soils chiefly of volcanic origin
and characterized also by heavy rainfall; and third, the vegetation
of the south shore, characterized by scanty rainfall with frequent
droughts, but occupying soils of diverse nature and origin.
In the central region but five major associations are represented
and they show no indication of successional relationships. In the
two coastal regions numerous associations occur, and these have
been grouped into series according to their successional relations.
While all of these series progress toward a single climax on the
north coast and toward another on the south side of the island,
they differ greatly in their origin. They are accordingly dis-
tinguished as hydrarch, halarch, xerarch, and mesarch series,
characterized respectively by an excess of water, an excess of salt,
a deficiency of water, or average environmental conditions in
their earliest stages.
Throughout the island numerous secondary associations are
present, in which the vegetation shows distinct changes due to the
activities of man. In some of these types the influence of mnan has
been confined to cutting of wood for construction or fuel, leaving
the vegetation in a relatively natural state; these we have dis-
cussed so far as time was available for their study. In other
secondary associations the original vegetation has been nearly or
completely destroyed, leading to the development, chiefly in waste
land or along roadsides, of entirely new assemblages of plants.
These we have omitted entirely.
The extraordinary development of agriculture in Porto Rico,
accompanying its dense rural population of nearly 300 per square
mile, necessarily interferes considerably with the completeness,
and possibly to some extent with the accuracy, of an ecological


survey. Several types of vegetation have been observed in a
natural or semi-natural state in only a single locality, similar
habitats elsewhere, so far as they have come to our attention, being
under cultivation. In such cases we infer that the original vegeta-
tion was uniform throughout and present a description of the
vegetation in the single station as typical of the whole extent in its
primitive condition. In other instances, the whole area of some
habitats, characterized by certain conditions of soil, moisture, or
other environmental factors known to be effective in determining
vegetation, is under cultivation, and we can only refer to its natural
vegetation as extinct.

During Tertiary time the northern portion of Porto Rico, lying
north of an irregular line extending approximately through Loiza,
Rio Piedras, Corozal, Ciales, Lares, and Aguadilla, was submerged,
and was covered by a thick deposit of limestone. Since that
time it has been subject to considerable fluctuations in level, so
that the sea has alternately receded from and encroached upon the
present land surface. Vegetation must have been established over
all of this area as rapidly as it appeared above the surface of the
water, and each subsequent extension or restriction of the land
surface, each recession or advance of the shore line, must have
been followed pari passu by significant changes in the distribution
of the various types of vegetation. What this old vegetation may
have been, or what changes in its composition or distribution it may
have suffered, must remain subjects for future investigation. But
it is clear that the processes or erosion and soil formation, initiated
immediately following the first emergence of the land, are still
in operation today, and that their progress, in some instances
hastened and in others retarded by the presence of vegetation, is
still influencing the plant life.
The highest part of the coastal plain now stands at an elevation
of about 400 meters (1300 ft.) above sea level and has doubtless had
a longer uninterrupted history as a land surface than any of the
lower portions. Vegetation here apparently reached its complete
development long ago as a mesophytic forest, but this forest has
been so nearly completely removed and the remnants so badly
altered by cutting that its nature can only be approximated. This
forest marked a stage of temporary climax in the vegetational
history, and from it a mesarch series (see p. 32) of successions
(Fig. 2) may be imperfectly traced, culminating in the climax
forest of the lowlands.
Although the level of the land has fluctuated during post-Tertiary
times, the general effect has been toward emergence of the land
and recession of the coast line, so that new land has been made
available to vegetation. Depending on the location of the coast


Mesophytic Forest Mesophytic Forest
Limestone Hills San Juan Dunes San Juan Dune



Thicketes Cocci
White Sand Dune Thicketse
Forest Du

/I Beach Thickets


Mudbank Typha-
Association > Mariscus



Fresh Water

"' Beach & Dune
Borrichia- Sand

Coccolobis L- Philoxerus
nes & Beaches

Playa Land Forest

Acrostichum F s
Mangrove Forest

Salt Water




FIG. 2. The principal associations of the northern coastal plain of Porto Rico, with their successional relations.








line, the environment of this newly exposed land has been hydrophy-
tic, at the edge of lagoons and along river banks; halophytic, in
salt marshes and mangrove swamps; or xerophytic, along sandy
beaches and on sand dunes. Correlated with these, three other
successional series may be traced, hydrarch, halarch, or xerarch
in their respective natures, and all culminate eventually in the same
climax forest of the lowlands. Since the climax forest has all been
displaced by agricultural lands, the complete series can in no case
be observed, and most of the intermediate stages have also been
profoundly affected by human influence. Furthermore, periods of
subsidence have from time to time tended to reverse the usual
successional series, but we have no present means of ascertaining
whether the intermediate stages have remained the same for both
emergence and subsidence. At the present time, a comparatively
recent period of subsidence, perhaps not yet completed, is still
exerting an influence on successions, and in many places a reversion
of vegetation toward its initial stage is now in progress. During
the last century the demands of a dense population for fuel, building
material, and grazing land for domestic animals have also made
noteworthy changes in the plant life, and have led to various
secondary successions.

The limestone strata of the northern coastal plain are organized
into at least five geological series, of which the Quebradillas and the
Cibao show comparatively little erosion, other than a gently rolling
surface and the usual steep-sided ravines with water courses at the
bottom. Over these strata almost all of the land has been claimed
for agriculture. On the Arecibo, Los Puertos, and Lares series, on
the other hand, in which the limestone is more easily permeable, an
extraordinary amount of erosion has reduced the original mass to an
area of characteristic hills, variously known as pepinos, mogotes, or
haystack hills. It is the opinion of geologists that most of this
erosion has been subterranean.
The hills are rounded to elliptic in general outline (Pl. 1) with
steep or often precipitous slopes, especially on the western exposure,
and sharp or ridge-like summits. Their bases are sometimes
separated by small valleys drained only by sink-holes at their
centers, sometimes in contact, and sometimes confluent for a portion


of their heights. Hundreds of such hills, ranging in height from
50 to 300 feet, are conspicuous features of the landscape from San
Juan to.Arecibo, and other belts of them, on the older strata of the
Los Puertos and Lares series, appear between Arecibo and Lares.
The outcrops of the limestone are heavily eroded (Pl. 2) into
sharp points, knife-like edges, deep crevices, and rounded pockets,
offering ample opportunity for the accumulation of vegetable
debris and the gradual development of a rich soil. Although the
soil mantle is everywhere thin and much of the rock is exposed, its
eroded surface and deep crevices provide sufficient root-space for
a dense vegetation and furnish anchorage for large trees. In the
valleys between the hills, along their bases, and on the narrow
ridges connecting them, the deposit of soil is naturally deeper, and
these areas have been extensively utilized for agriculture.
The soil of the small intervening valleys or depressions is some-
times alkaline, but in many cases the lime has been leached out or
the alkalinity overcome by the accumulations of humus. The
crops in these valleys are extremely varied. Sugar cane is grown
to some extent and a little coffee is found but it does not thrive as
well as on the volcanic soils of the higher altitudes. The most
prominent crop at the present time is the citrus, which thrives in
these, soils and also on the sandy soils nearer the coast. Pine-
apples are also grown extensively but do not succeed in the presence
of lime unless there is a large amount of humus. In some of the
valleys the pineapples show a decided chlorosis, while in others
there is very little or none. Bananas are found in the protected
places but are not grown as a commercial crop. Tobacco is
grown to some extent but is of very little importance.
Both surface and subterranean erosion are still active. Rain
water travels but a short distance over the surface before sinking
into the numerous crevices and pits, which are accordingly being
constantly deepened, widened, and kept filled with soil. Caves
are frequent and sometimes of large size. Recently fallen blocks
of limestone may be seen at the base of most of the cliffs. The
intervening valleys are not drained by surface streams, and their
catchment water escapes through sink-holes at their deepest part.
Most of these are filled with soil, but open sink-holes of unknown
depth are frequent and often of large size, and the noise of subter-
ranean streams may be heard at the bottom of a few of them toward
the western end of the island. So complete is the subterranean


drainage that streams of local origin scarcely exist through the
limestone belt, although it is crossed by the larger rivers which rise
in the mountains and flow northward to the sea. Even some of
these have not eroded a normal channel but flow for a part of their
course underground. Long-continued erosion has reduced some of
the hills to mere fragments of rock, and the broad plains which
separate the easternmost hills have been explained as representing
the completion of the process and the reduction of the whole to base
Although the rainfall of the limestone belt decreases somewhat
from east to west, it is always sufficient to support a mesophytic
forest, and the nature of the present plant covering supports the
belief that the normal vegetation was originally of that type.
Close to and paralleling the north shore, a strip of low limestone
hills known as the San Juan formation stretches with occasional
interruptions from Camuy to a short distance east of San Juan.
They are composed of a calcareous sand, bound together by an
organic cement derived from vegetable decomposition, and origin-
ated from a series of fixed dunes in comparatively recent times.
The strip seldom surpasses a kilometer in width and the highest
elevations probably do not exceed 50 meters (150 ft.). Here
subterranean erosion is unimportant, but the easily weathered
surface rapidly breaks down into sand. The original sharp dune
topography has accordingly been softened and the underlying rock
is almost completely hidden beneath a blanket of soil. Practically
all of this formation has been placed under cultivation, but the
few thickets which still remain resemble those of the limestone
hills so closely that it seems probable that the original vegetation
of the two was essentially identical.

The mesophytic forest which originally covered the limestone
hills and the San Juan consolidated dunes has been either entirely
destroyed or seriously modified throughout its whole extent. Its
ecological nature can not even be approximated from the few
mature trees left standing, since they are invariably species of no
economic value, such as Elaphrium Simaruba (L.) Rose, Clusia
rosea Jacq., Ficus laevigata Vahl, Ficus Stahlii Warb., and
Gaussia attenuata (Cook) Beccari. It seems probable that the
same type of forest, doubtless with numerous component species


and a rich mesophytic ground flora, extended over the hills from
their tops to the bottom of the intervening valleys, and occupied
also the smoother uplands and ravines of the Cibao and Quebradil-
las formations.
Murphy (25) in his discussion of the forests of Porto Rico, also
refers to the extinction of this type of forest, and states that Coc-
colobis grandifolia Jacq., Simaruba Tulae Urban, Petitia domingensis
Jacq., Amomis caryophyllata (Jacq.) Krug & Urban, and Buchenavia
capitata (Vahl) Eichl. are reported to have been formerly common
on the limestone hills, together with other species of large trees.
Such a forest marked the close of a successional series which must
have begun soon after the emergence of these limestones above the
sea, passed through unknown intermediate stages in its develop-
ment, and existed as a temporary climax for thousands of years.
Keeping pace in its growth and regeneration with the erosion of the
hills, it held its position continuously, disappearing only when
certain of the hills were completely eroded and reduced to base
level, and interrupted only when falling ledges exposed new sur-
faces of fresh rock. Even these new surfaces were re-occupied as
rapidly as accumulating soil furnished sufficient root-space for the
With the great development of population during the last century,
occupying and using for agriculture all the level land near the lime-
stone hills and all of the comparatively level uplands of the Cibao
and Quebradillas limestones, there began a serious modification of
the mesophytic forest, resulting in a secondary succession to the
present second-growth thicket association. This modification was
caused by cutting all the larger trees, first probably for construction
purposes and later for firewood, and has affected all the limestone
hills and the ravines and bluffs of the Cibao and Quebradillas
uplands. Clearing has been followed in some places by using the
hills for pasture for domestic animals, particularly near the villages,
and in some instances houses have been built on the hills and garden-
ing attempted on a small scale.
These activities have made a great change in the environment,
greatly increasing the amount of light available to the ground
flora, increasing transpiration by greater exposure to the wind,
reducing the thickness of soil by furthering surface erosion, hinder-


ing the accumulation of vegetable mold, and reducing the amount
of soil water by facilitating the run-off and increasing surface
evaporation. While they have probably not led directly to the
extinction of many species, they have undoubtedly changed their
numerical proportions greatly, decreasing the numbers of those
less adapted and increasing those better adapted to the changed
environment; they have changed the aspect of the vegetation from
a forest to a thicket of shrubs, and they have favored the establish-
ment of numerous additional species which have been quick to
immigrate from surrounding regions. Neither have they yet
produced their maximum effect. With a continually increasing
population, wood is cut for fuel or charcoal at smaller sizes, gradually
reducing the density of the shrubby vegetation, and the foraging of
goats will doubtless eventually cause the complete extermination
of certain species. The hills nearest the numerous towns and
barrios are regularly more sparsely vegetated, with greater sur-
faces of bare rock exposed, and we may confidently expect a
similar destruction of vegetation over ever greater areas as time
goes on.
In the average development of the second-growth thickets at
the present time, the hills still appear from a distance to be forested
(Pl. 1), but the apparent forest cover is only a dense jungle of
shrubs. Rising above the shrubs are isolated trees of various
worthless species. The tall, exceedingly slender stems of Gaussia
attenuata (Cook) Beccari, endemic to these hills and to similar
habitats on the southwestern coast, appear as scattered individuals
and small groups, usually at the very summit of the hills. From
a little distance the trunks are lost to sight and the cluster of leaves,
usually only five or six in number, apparently floats in the air.
The widely branched, crooked boles of Elaphrium Simaruba (L.)
Rose are always common and are especially conspicuous in their
leafless period, which extends through a considerable portion of the
year. The coarse, ungainly stems of Cecropia peltata L. are
conspicuous in a wind, when the white lower surface of the leaves
is exposed. The densely leafy crowns of Ficus Stahlii Warb., Ficus
laevigata Vahl, and Clusia rosea Jacq. attract less attention except
when growing on the profile of the hill.
At the base of a hill, the cultivated land is bordered with a
narrow zone of weedy species, in which Piper aduncum L. is most
abundant and characteristic. Associated with it are Miconia


laevigata (L.) DC., Duggena hirsuta (Jacq.) Britton, and Critonia
portoricensis (Urban) Britton & Wilson, while Pothomorphe
peltata (L.) Miq. is common on the moister north slopes. The
herbaceous plants here are mostly weedy species, such as Elephan-
topus mollis HBK., Bidens pilosa L., and Osmia odorata (L.) Sch.
Above this zone, a dense thicket (Pl. 2) of shrubs and small
trees extends to the top of the hill. Their height is usually from
3 to 5 meters (10-15 ft.), although occasionally better specimens
may be found in places not recently visited by the woodcutters.
The assemblage is composed of a large number of species, mostly
growing so scattered that no one or no group of species appears
to be predominant. So far as our observations could determine,
the most abundant species in number of individuals are Quararibaea
turbinata (Sw.) Poir., Piper Amalago L., Nectandra patens (Sw.)
Griseb., Psychotria pubescens Sw., Picramnia pentandra Sw.,
Acalypha portoricensis Muell. Arg., Zanthoxylum martinicense
(Lam.) DC., Eugenia procera (Sw.) Poir., and Trichilia pallida Sw.
Of these only Zanthoxylum martinicense (Lam.) DC. is at all
conspicuous, because of the stout conical thorns on its trunk.
Various other species are much more striking and conspicuous
because of their habit or vegetative form. Trees or tall shrubs of
Clusia rosea Jacq. are common, and send out long aerial roots of
large size along the face of the cliffs. The smooth glossy leaves,
with unusual venation, of Calophyllum antillanum Britton, Manil-
kara nitida (Sess6 & Mog.) Dubard, Amomis caryophyllata (Jacq.)
Krug & Urban, and Rheedia acuminata (Spreng.) Tr. & Pl. are
conspicuous. Holly-like leaves with spinose teeth occur on
Malpighia coccigera L. and Drypetes ilicifolia Krug & Urban,
while somewhat similar spiny leaves are found on the sumac-like
poisonous Comocladia glabra (Schult.) Spreng. The broad hairy
leaves of Ochroma pyramidale (Cav.) Urban, the peculiarly lobed
leaves of Curcas hernandifolius (Vent.) Britton, and the crowded
compound leaves with broad leaflets of Dendropanax arboreum
(L.) Dene. & Pl. give them aspects quite different from most other
species. Occasional plants are found of the prickly stemmed
palm, Bactris acanthophylla Mart., while the spiny stems of Antha-
canthus spinosus (Jacq.) Nees are often noteworthy. The gesneriad
Pentarhaphia albiflora Dcne. grows in crevices of the cliffs and
attains a height of one to two meters, forming a widely branching
shrub with ascending stems.


The whole thicket is interlaced with the tangled stems of numer-
ous lianas. Among these the most conspicuous are Lasiacis
sorghoidea (Desv.) H. & C., frequently forming dense and almost
impenetrable masses, and the stout, woody, thorny stems of Acacia
riparia HBK. Tragia volubilis L., its stems and leaves armed
with stinging hairs, is common. A species of Hippocratea, probably
H, volubilis L., is also abundant and is interesting for the peculiar
morphology of its tendrils. Small plants of Batocydia Unguis
(L.) Mart. are frequent on the trunks of trees or the larger shrubs.
Among other lianas there may be noted Stigmaphyllon tomentosum
(Desf.) Ndz., Elsota virgata (Sw.) Kuntze, Smilax coriacea Spreng.,
Cissampelos Pareira L., Serjania polyphylla (L.) Radlk., Exogonium
repandum (Jacq.) Choisy, and two or three species of Ipomoea.
Epiphytes, which were probably abundant in the original forest,
are comparatively few in numbers, and are now usually to be found
on exposed ledges of rock. The most conspicuous is Anthurium
acaule (Jacq.) Schott, its leaves a meter long and forming a large
rosette. Long stems of Philodendron Krebsii Schott ascend the
few trees or droop over the cliffs. Vanilla Eggersii Rolfe, with
thickened stems and small leaves, has a similar habit. Slender
plants of Marcgravia in their juvenile stage are appressed closely
against shady rocks. Banks of Pitcairnia angustifolia (Sw.)
Redoute are massed on rock ledges, and various other bromeliads,
usually observed only in sterile condition, are attached to the
trees of Ficus and Clusia.
Herbaceous plants occupy a distinctly subordinate position, both
in number of species and in quantity of individuals. The largest
is Alpinia aromatica Aubl., reaching a height of more than two
meters. The most conspicuous is Zamia latifoliolata Preneloup,
which is widely distributed, and frequently becomes abundant
after clearing. The shade-loving grasses Ichnanthus pallens (Sw.)
Munro and Pharus glaber HBK. are common. The scarlet flowers
of the delicate Ruellia coccinea (L.) Vahl make it conspicuous when
in bloom. Pilea microphylla (L.) Liebm., Pilea nummulariaefolia
(Sw.) Wedd., Peperomia rotundifolia (L.) HBK., and P. magnoliae-
folia (Jacq.) A. Dietr. occur on shaded cliffs. Polypodium phyl-
litidis L., resembling an Anthurium in general habit, is the most
noteworthy fern, although Adiantum cristatum L. is more abundant,
and various other species occur more rarely.
This type of vegetation is maintained essentially unchanged over


all the scattered hills and over the solid limestone bluffs of the
Quebradillas formation. There is in general less variation in
vegetation from one hill to another than from the north to the
south slopes of the same hill. The limestone strata dip about 5
degrees to the north, and, while the limestone is not prominently
bedded, this slight inclination probably helps maintain a better
supply of moisture on the north side, which also has a somewhat
less direct exposure to the sun during most of the year. On the
north side of a hill the shrubbery is denser, bromeliads, aroids,
ferns, and Ruellia coccinea (L.) Vahl are more abundant, the
accumulation of leaf mold is conspicuously deeper, and the whole
aspect is distinctly mesophytic. On the southern exposure the
shrubbery is more open and more tangled with Lasiacis and Acacia
riparia HBK., microphyllous and other semi-xerophytic species
are more abundant, and the soil is obviously thinner and drier.
Exposed cliffs and ledges on the south side are nearly devoid of
plants, except trailing vines hanging from above, while even high
cliffs on the north side are usually masked behind a dense growth
of lianas, epiphytes growing on the rock, and herbaceous plants
in the numerous ledges and pockets.
Under the present conditions of continued human interference,
the trend of the vegetation is toward still further reduction in
density to a mere rocky hill, occupied chiefly by weedy species not
relished by domestic animals.

Along the north side of the belt of limestone hills, the hills usually
stand farther apart and are separated by level or gently undulating
plains of a red clay-loam soil. So far as known to us, these lands
are all under cultivation, mostly in orchards of oranges or grape-
fruit. The fact that the pineapple succeeds upon them indicates
that they are no longer calcareous in nature, although derived
directly from the disintegration of the adjacent hills. This is
doubtless due to the neutralization of excess of lime by the long-
continued deposition of humus. We can make no statements
concerning the nature of their original vegetation, but we believe
that the latter was not a member of the successional series begin-
ning with the forests of the limestone hills.


In several places along the north shore there is a belt of white
sands, lying between the ocean and the lower limestone hills, and
regarded as the result of the disintegration and base-leveling of the
hills. These sands appear on the soil survey of the Arecibo to
Ponce Area under the name of Arecibo Sand (17, 804-805). They
are perhaps most extensively developed between Manati and
Dorado, but also occur as far east as San Juan. They are almost
exclusively under cultivation, and the only development of natural
vegetation upon them, so far as known to us, has been preserved
by Miss Livingston on her estate west of Dorado. The following
description has been taken wholly from this location (Pl. 3).
The forest occupies essentially flat, level land adjacent to the
coast and but a few feet above sea level. A few small depressions
and some beds of intermittent streams exist within the area, and
extend probably nearly or quite to sea level. At the seaward side
there is a slight elevation, caused by the accumulation of wind-
borne sands and occupied by a narrow strip of beach thickets,
described below.
The soil is a white calcareous sand of unknown depth. The
surface is covered with a thin layer of fallen leaves and twigs,
rarely more than 1-3 cm. deep, and decaying beneath into a very
thin deposit of humus. Below this the sand is stained dark brown
by decayed vegetable matter; this color becomes gradually paler
with increasing depth, and is almost white at two or three deci-
meters below the surface, with no distinction between the surface
soil and the subsoil. This upper layer of soil is crowded with a
mass of feeding roots while the lower levels have much fewer.
The surface is completely shaded, moisture is well conserved, the
trade winds of the shore do not penetrate more than 50 meters,
and conditions are generally favorable for the development of
a high type of mesophytic forest.
We did not observe the contact of this association with others on
the landward side, where other tracts of the same white sands are
all under cultivation.
The general height of the dominant trees is now about 20 meters
(60 ft.). In the forest there are comparatively few veteran trees,
and it seems to have been culled for lumber from time to time, but
a general clearing has probably never taken place. These veteran
trees are supported by great numbers of young trees of all ages and


sizes, ranging down to seedlings less than a year old. In comparison
with a hardwood forest of the United States, there are fewer
veterans and seedlings, and far more saplings. The forest floor
is accordingly less densely covered but the general density of
growth appears heavier, because of the numerous saplings of the
dominant trees. Shrubs are in the minority and herbaceous vege-
tation is almost lacking.
Among the trees, two species are especially noteworthy for their
size and abundance. Mammea americana L. and Calophyllum
antillanum Britton both reach heights of 20 meters (60 ft.) with
trunk diameters of nearly or quite one meter. Large specimens of
both species are always within sight, and usually several of each.
Saplings of both species are abundant, and their seedlings form
possibly half of the ground flora. It is impossible to range the
other trees in order of their abundance but certain species can be
classed as generally common. Dendropanax arboreum (L.) Dene.
& Pl. is nearly ubiquitous; it seldom exceeds 10 meters (30 ft.) in
height, or its slender trunks 2 dm. (8 inches) in diameter. Clusia
rosea Jacq. is conspicuous because of the presence of long stout
aerial roots and the color of its bark. Elaphrium Simaruba (L.)
Rose is abundant. Amomis caryophyllata (Jacq.) Krug & Urban
becomes 20 meters (60 ft.) high. The low wet places are often
occupied by groups of Roystonea borinquena Cook. Hundreds of
small saplings of Jambos Jambos (L.) Millsp. occur in thickets
along some of the trails. Of numerous other species of trees which
compose the forest, we may mention Laugeria resinosa Vahl,
Zanthoxylum martinicense (Lam.) DC., Gymnanthes lucida Sw.,
Nectandra coriacea (Sw.) Griseb., Manilkara duplicate (Sess6 &
Mog.) Dubard, Diospyros ebenaster Retz., Enallagma latifolia (Mill.)
Small, and Acrocomia aculeata (Jacq.) Lodd.
As has been mentioned, the majority of the lower woody plants
consist of seedlings of the larger trees, including several species
whose presence in the forest layer was not noted. Of the true
shrubs, the single abundant species is Chiococca alba (L.) Hitchc.
Second in abundance is Piper citrifolium Lam., and among
several other species there may be noted Parathesis serrulata
(Sw.) Mez, Psychotria undata Jacq., Eugenia monticola (Sw.) DC.,
Gyminda latifolia (Sw.) Urban, Casearia guianensis (Aubl.) Urban,
and Piper Amalago L. Lianas are very few in number. The
thorny stems of Pisonia aculeata L. climb to a height of 6 meters


(20 ft.) and an unidentifiable species reaches a diameter of 5 cm.,
looping from one tree to another to a height of 15 meters (50 ft.)
with a much greater actual length. Rourea surinamensis Miq.
becomes 6 m. high. A few plants occur of Vanilla Eggersii Rolfe,
Batocydia Unguis (L.) Mart., Smilax coriacea Spreng., and Stig-
maphyllon tomentosum (Desv.) Ndz., while small sterile plants of
Marcgravia are rare. Epiphytes are also few. Sterile bromeliads
of unidentified species occur in some of the trees; the ferns Poly-
podium phyllitidis L. and P. lycopodioides L. are occasional, and
the small orchid lonopsis satyroides (Sw.) Rchb.f. is rare.
Herbaceous plants are so few and so widely scattered that they
play a very subordinate part in the vegetation. The low grasses
Pharus glaber HBK. and Ichnanthus pallens (Sw.) Munro are the
most abundant. Nephrolepis exaltata (L.) Schott is occasional.
Two orchids were noted, Beadlea cranichoides (Griseb.) Small and
Liparis elata Lindl.
In the rather numerous depressions a thin layer of wet muck,
composed of decaying vegetable matter overlies the sand, but the
only change in the forest flora is seen in the greater numbers of
Roystonea borinquena Cook and the frequent thickets of Jambos
Jambos (L.) Millsp. In a few places these depressions are deep
enough to harbor permanent standing water in small muddy pools,
which may be occupied by small clumps of Typha angustifolia L.
Throughout the whole association there is no indication of
physiographic changes now in progress, except for the presence of
the small depressions, or of vegetational processes which might
lead to further changes in the flora. It is not certain, however,
that this is the true climax association of the region, since further
removal of the lime by continued leaching and further accumula-
tion of humus may conceivably result in a succession. If so, the
following vegetation might well be the same as the forest, now
extinct, which originally occupied the alluvial playas of the same
region, and which we regard accordingly as the true climatic climax

Just as the normal forest of the limestone hills has been converted
into a thicket vegetation by human activities, so has the forest of
the white sand suffered similarly and been reduced to a similar
jungle of shrubs and young trees.


The best development of this vegetation, from which the follow-
ing description is made, occurs along the south shore of Laguna
Tortuguero. Here the surface is again level, and frequently
broken by small shallow ravines occupied by intermittent streams
during the wet season of the year. Long exposure to wind and sun
and a great reduction in the annual accession of fresh vegetable
matter have furthered the decomposition of the surface humus, and
reduced the soil to a pure white sand. This lowers the water-
capacity of the soil, favors surface evaporation, hastens the run-
off of rain-water and tends to produce a distinctly xerophytic
condition which is reflected in the plant life.
In general appearance, the vegetation consists of alternating
areas of open places and dense thickets (Pl. 4). The latter, which
generally follow the shallow ravines and spread laterally from them,
have an average height of 3 to 5 meters (10-15 ft.) and are almost
impenetrable. The open places, occurring consequently on the
level ground between the ravines, are occupied by a sparse growth
of herbs and low shrubs and much bare sand is exposed.
The most abundant shrub, probably outnumbering all the
others together, is Chrysobalanus Icaco L. It forms a bushy
growth, the lower branches usually resting on or developed close
to the ground, and is gregarious, several plants of different ages
growing close together. Fruit is borne when the plants are only
a meter high, but large examples attain a height of 5 meters (15
ft.). Second in abundance is probably Byrsonima spicata (Cav.)
DC., growing tall, erect, and rather slender, while Myrcia splendens
(Sw.) DC., a bushy shrub with glossy dark green foliage, is almost
equally plentiful. Other shrubs which contribute prominently to
the thickets are Randia mitis L., Calophyllum antillanum Britton,
Miconia racemosa (Aubl.) DC., Miconia prasina (Sw.) DC.,
Ocotea leucoxylon (Sw.) Mez, Ecastophyllum Ecastophyllum (L.)
Britton, Casearia sylvestris Sw., Myrica cerifera L., Cupania
triquetra A. Rich., Acrodiclidium salicifolium (Sw.) Griseb., Myrcia
citrifolia (Aubl.) Urban, and Taonabo peduncularis (DC.) Britton.
All of these have the same bushy form and about the same height.
Didymopanax Morototoni (Aubl.) Dcne. & Pl. is not common, but
is conspicuous because of its large leaves with bronzed lower
surface. Ouratea littoralis Urban is frequent and conspicuous
when in bloom. Occasional thickets of Acrocomia aculeata (Jacq.)
Lodd. occur.


Lianas are neither abundant nor distinctive. Smilax coriacea
Spreng., Elsota virgata (Sw.) Kuntze, Cissampelos Pareira L., and
Cissus erosa L. C. Rich. are the most conspicuous. Epiphytes
are absent. Herbaceous plants are absent under the thickets of
the uplands, but in the moister places at the bottom of the small
ravines scattered plants of Nepsera aquatica (Aubl.) Naud., Xyris
Elliottii Chapm., and Setiscapella subulata (L.) Barnh. occur.
On the flat open places between the thickets, the characteristic
vegetation is a sparse growth of Mitracarpus portoricensis Urban
and Ascyrum hypericoides L., both bushy-branched half-shrubs 3 to
6 dm. (1-2 ft.) high. Associated with them are scattered mats of
Portulaca pilosa L., isolated low plants of Crotalaria retusa L.,
Chamaecrista diphylla (L.) Greene, and C. mirabilis Pollard, and,
rarely, patches of Cladonia rangiferina L.
There is reason to believe that this area has been cut over for
fuel frequently in the past, and at the time of our second visit
cutting had begun again. A tract of two or three acres had already
been cleared and the wood prepared for charcoal burning. In
the frequent cutting we may discover an apparent reason for the
presence of so many species not noted in the original forest of the
same habitat. The change in environment has been so serious
that most species of the original forest have not survived. Their
places have been filled partly by the more abundant development
of those species which could endure the new conditions and partly
by the immigration of additional species from other habitats.
Among these Chrysobalanus Icaco L. has had the advantage of
proximity, of complete adaptation to a similar xerophytic habitat
in the calcareous sands of the seashore, and of efficient means for
rapid migration through its edible fruits. It has accordingly
become the dominant shrub.
The history of the thickets will probably close in the near future
by the use of the land for agriculture; left to itself, the gradual
accumulation of humus, the conservation of soil-moisture, and the
decrease of the light might eventually lead to the re-establishment
of a normal forest.

Recent changes in the elevation of the coast line have brought
the sea in direct contact with some of the consolidated sand dunes
of the San Juan Formation. As a result the hills are being rapidly


eroded, and the second-growth thickets which cover them are
passing through a reversionary successional series in which the
vegetation is ultimately completely destroyed. While these
successions are logically a part of the mesarch series, they also
portray a possible stage in the history of the xerarch series, and
the vegetation associated with them is closely akin to that of the
coastal sand dunes in its environment, dynamics, and specific
composition. We have accordingly taken up their description
under the xerarch series.
The limestone deposits originally covered most of the area of the
northern coastal plain and their uneroded portions developed a
mesophytic upland forest which has now been destroyed and re-
placed by the present thickets of the limestone hills. Erosion has
reduced much of the limestone strata to base level, and the final
evidence of their former occurrence is seen in the level areas of
white calcareous sands. These lowland sands support normally a
mesophytic forest which has been almost entirely destroyed and
replaced by agriculture. When not cultivated, but merely cut
severely and repeatedly for wood, the forest is succeeded by a
thicket association of different specific composition.

Subseries A. Beaches and Beach Thickets
Along the whole of the shore of the northern coastal plain, and
also beyond the limits of the Tertiary formations to the east and
south along the eastern end of the island approximately to Fajardo
Playa, the free exposure to the almost constant trade winds and
surf produces ideal conditions for the development of sand dunes.
But in the sheltered coves to the west of headlands of the San
Juan formation, where the force of the trade winds is greatly
reduced, and on the northeastern part of the island, where an
adequate supply of sand seems to be not available, dunes have not
been produced. Under such circumstances the sandy beach rises
gently to a height of one to two meters (3-6 ft.) above the water
and then passes without change of level into flat sandy plains or
comes immediately in contact with the hills of the volcanic rocks
of Cretaceous age. In each case the vegetation exhibits three zones,


a beach association of scattered plants of herbaceous species, a
thicket of semi-xerophytic shrubs, and a third, beyond or above
the influence of maritime conditions, now generally destroyed and
replaced by agricultural lands or pasture. There is reason to
believe that the level sandy plains of maritime origin were originally
occupied by a forest similar to that of the white sands but succeed-
ing the beach thickets, while the more ancient hills bore another
type of forest of entirely different genetic history, merely in geo-
graphic contact with, but unrelated to the thickets.

The strip of exposed beach is usually from 5 to 10 meters (15-
30 ft.) wide between the water and the edge of the thickets. It
supports a very sparse vegetation of herbaceous species. All of
these are subject to destruction at irregular intervals by waves of
greater than normal height, so that the flora consists of species
quickly established from seed, or of others which root at the edge
of the thickets and trail down over the surface. Prominent in the
latter class are Ipomoea Pes-caprae (L.) Roth, Canavali maritima
(Aubl.) Thou., and Sporobolus virginicus (L.) Kunth. Among the
former, the more abundant species are Fimbristylis spathacea
Roth., Kyllinga pervuiana Lam., Remirea maritima Aubl., Cakile
lanceolata (Willd.) 0. E. Schulz, Chamaesyce buxifolia (Lam.)
Small, Borreria verticillata (L.) Meyer, and Diodia maritima Thonn.
These plants all represent the continuous attempt of vegeta-
tion to obtain possession of a strip of land where a permanent
vegetation is impossible. They are destroyed by every storm and
appear anew shortly afterward, the location and the number of
individuals of each species depending entirely on the accidents of
seed dispersal.
The lower level of the beach thickets marks the upper limit of
ordinary storm-wave activity. At this shoreward edge, the most
important pioneer shrub is Coccolobis uvifera (L.) Jacq., which is
represented by more individuals than all the others together.
Chrysobalanus Icaco L. is second in abundance and usually some-
what farther back from the shore. Other species associated with
it include Scaevola Plumierii (L.) Vahl, Lantana involucrata L.,
Randia mitis L., and Guilandina Crista (L.) Small. All the herbace-


ous species of the beach are present, together with Crotalaria
retusa L., Indigofera suffruticosa Mill., Stylosanthes hamata (L.)
Taubert, and Bidens pilosa L., the last forming dense mats.
The structure of the wider beach thickets is excellently illustrated
at km. 36, east of Mameyes. Along the shoreward margin at this
point the thickets seldom attain a meter in height, but each meter
back from the margin shows an increase in the height of the plants,
in the number of individuals, and in the number of species. The
thicket becomes dense and almost impassable, with the plants
reaching heights as great as 5 meters (15 ft.). Erithalis fruticosa
L. becomes very abundant; other characteristic species are Eugenia
axillaris (Sw.) Willd., Rauwolfia tetraphylla L., Ernodea littoralis
Sw., Citharexylum fruticosum L., Jacquinia Barbasco (Loefl.) Mez,
and Psychotria undata Jacq. The slender branches of Plumiera
alba L. project above the general level, and the whole thicket is
interlaced by tangled masses of Guilandina Crista (L.) Small,
Smilax coriacea Spreng., Serjania polyphylla (L.) Radlk., and
Distictis lactiflora (Vahl) DC. Accompanying this greater density,
the pioneer shrubs, Coccolobis, Chrysobalanus, and Scaevola, dis-
appear completely, and at a distance of 50 meters (55 yards) from
the ocean trees of Tabebuia pallida Miers, Elaphrium Simaruba (L.)
Rose, and Sideroxylon foetidissimum Jacq. indicate the completion
of the transition from a beach to a forest environment. Farther
from the shore the original vegetation has been replaced by pasture,
but the arborescent species still represented make it appear probable
that the present pasture land was occupied by a mesophytic
forest, and we believe that it was essentially like that of the white
sands at Dorado, previously described.
At Dorado, the narrow beach is bordered by a very narrow zone
of beach thickets, in which Coccolobis uvifera (L.) Jacq. is again the
dominant plant. This is followed almost immediately by the
mesophytic forest, but several thicket species, such as Rauwolfia
tetraphylla L. and Erithalis fruticosa L., are prominent near the
shore. At Mameyes, the presence of young trees of Tabebuia
pallida Miers near the front of the thickets may indicate an at-
tempt of the forest to establish itself closer to the shore.
The sand seems to be completely stabilized as far as the front
of the thickets, and the environmental change between the thicket
zone and the forest at its rear seems to consist almost entirely of a
difference in wind. Each shrub rises slightly higher than those


in front of it, and the general level slopes up from a height of a
meter or less to that of the mature forest. As young trees establish
themselves farther forward under the shelter of the thickets, the
pitch of the thicket zone becomes steeper and its width accordingly
narrower, until in its completed stage, as at Dorado, the meso-
phytic forest comes almost to the beach. At Mameyes, the removal
of the forest and the frequent cutting of young trees for fuel has
served to maintain the gradual pitch of the surface and the greater
width of the thicket zone.
It is probable that these beach thickets are very like the dune
thickets, described below, in their component species, although
leading to a different successional result. It may also be noted
that at Fajardo Playa a type of thickets occur which are inter-
mediate in structure between the normal thickets just discussed
and a mangrove swamp. This occupies a habitat intermediate in
environment between the coastal swamps and the sand beaches,
and will be discussed later.

Subseries B. Sand Dune Vegetation
The sand dunes of Porto Rico present two very different aspects
along the north coast, depending on their age. The youngest
series, of recent development, consists of active dunes of loose
calcareous sand, exhibiting typical dune structure and freely
moved by the wind, except where it is held by a mantle of vegeta-
tion. The oldest series consists of larger dunes of ancient origin,
in which the sand has been consolidated by an organic cement into
a limestone rock, known as the San Juan formation. Its name is
derived from the city of San Juan, which is built on hills of this
type. Sand dunes of one or both of these series extend along the
shore, with some small interruptions, from Camuy to somewhat
east of San Juan.
The dunes of the San Juan formation are developed in two, or
possibly three, parallel series. The inland, older series averages
approximately a kilometer in width and rises usually to a height of
20 to 50 meters (60-160 ft.) above the sea. It presents the usual
sweeping, rounded contours of dunes, but with somewhat more
flattened crests and more gently sloping sides, due to long-continued
erosion from the summits and deposition in the valleys. The
outer series is much narrower and lower and its steeper slopes and
sharper summits give evidence of its more recent formation. This


outer series is now in contact with the sea at many points and
exhibits undercutting and extensive erosion by the waves as a
result. At some places the waves have produced breaks through
the dunes (Pl. 11), allowing the sea to pour in toward the older
series farther inland; in other places the rocks have been reduced
to mere isolated crags barely emerging from the water and doomed
to early destruction, and in. still other locations the outer series
has been completely destroyed by the waves.
The newer series of active dunes is composed of sand washed
ashore by the waves and blown inland by the wind. Most of
this sand has apparently been derived in recent times from the
disintegration of the outer series of the San Juan formation, and as
a consequence lies between the two series of the older formation
and is best developed opposite breaks in the outer series. Where
both series of the San Juan formation are still intact, they are
separated only by a narrow valley with a fairly deep and fertile
The history of the dunes may accordingly be summarized as
1. At some past period of time the general level of the land was
lower than at present and the sea extended inland to the base of
the inner series of consolidated dunes. Large active dunes were
formed along the shore by the action of wind and wave and ulti-
mately consolidated into the present inner series of the San Juan
2. An elevation of the land caused a recession of the shore line
to the northward, and a new, but smaller, series of active dunes
was formed by the same action of wind and wave. This was then
consolidated into the present outer series of the San Juan formation.
3. A subsidence of the land brought the sea actively into contact
with the outer series, led to its partial destruction, and to the
formation of the present series of active dunes from the products
of disintegration.
We have no evidence by which future changes in the level of the
land may be predicted, but any considerable change will have an
immediate effect on the history of the dunes. If the land continues
to sink, the outer series of dunes will be completely destroyed and
the sea will begin to encroach on these of the inner series. If the
level remains constant, the destruction of the outer series will be
completed, and the present active dunes will eventually be con-


solidated into limestone. If, on the other hand, the land is again
elevated, the remnants of the outer series will be preserved, the
present active dunes will be consolidated, and a new series of
active dunes will appear along the new shoreline.
The recent dunes have at present no agricultural value, aside
from a few attempts to plant coconuts upon them. On the con-
trary, they frequently become an actual menace through their
encroachment on agricultural land or garden plots and their
covering of roads and railroads. Numerous small houses have
been built on the lee side of the dunes, with small plantings of
yautia, sweet potato, and tobacco. Such gardens usually occupy
the fertile valley soil, and are protected by the dunes from the
force of the steady winds from the sea. In some cases, small
plantings have been made along the crest of the outer series of
dunes of the San Juan formation, and there it has been found
necessary to build crude windbreaks of palm leaves or other mater-
ials. The inner consolidated dunes are almost entirely under
cultivation. Cane is planted on the deeper soils, while the thinner
soils are used for grazing, and a variety of crops are grown on a
small scale near the numerous houses. The intervening valley
land is generally in cane, except in places low enough to be brackish,
where coconuts are substituted.
The history of the dunes of both series is intimately connected
with their natural vegetation, and nearly all stages in the develop-
ment and later destruction of the plant life may be observed.
Vegetation seizes the sand as soon as it becomes quiet, and continu-
ally attempts to extend farther toward the sea until it is effectually
checked by the destructive action of wind and waves. The first
outposts of plant life are followed by thickets of shrubs as the sand
becomes stabilized. As the sand becomes further modified into
actual soil and mixed with organic matter from the decomposi-
tion of plant material, the thickets are followed by forests. When
the dunes are again attacked by the sea, the vegetational series is
reversed, the forests are replaced by thickets, and the thickets
in turn disappear, leaving the unprotected rock fully exposed to
the destructive influences of wind, wave, and rain.
The mechanics of dune formation has frequently been discussed
and needs but a brief recapitulation here. As the surface layers of
sand on the beach dry under the sun, the loose grains are blown
by the wind. The general effect of the wind, when free from other


influences, is to level the sand and not to produce dunes. But
any obstruction to the wind reduces its velocity and its carrying
power, and causes it to drop its load of sand at that point and to
produce a hillock of sand behind the obstruction. This miniature
dune cannot exceed the height of the obstruction which produced it.
But if the obstruction is a growing plant, able to rise continually
above the level of the dune, the height of the dune increases until the
force of the wind is no longer sufficient to raise sand to its top.
But the development of a dune is seldom so simple as this and
many conditions conspire to its modification or destruction. As
a dune grows higher, it also grows broader, and other plants are
needed to protect its lower slopes. The establishment of these
plants in a shifting sand is not easy, and dunes are frequently
destroyed by undercutting at the base on the windward side,
either by the wind or by waves. As the dune becomes higher, the
force of the wind at its top is increased and there is a growing
tendency for sand to blow over the top and fall down on the lee
side. Not all plants are equally fitted for holding the sand in
place at the top, and the dunes vary in size with the nature of the
plant which produced them. The highest active dunes along this
part of the coast are probably not more than fifteen meters high.
As the small dunes grow, they unite into a long dune ridge, which
may extend for a long distance parallel to the shore. Any break
in the plant covering of such a ridge, either natural or caused by
the activities of man, gives the wind direct access to the sand.
Unless such a gap is immediately closed, it is soon widened by the
wind, vegetation is destroyed more rapidly than it is produced,
great quantities of sand sweep through the opening, and a new
series of dunes is formed on the lee side of the original one.
In general aspect the active dunes are exceedingly varied. In
fact, no two areas of them seem to agree, either in size, in topo-
graphy, in the effects of wind action, or in vegetation. The beach
may be broad or narrow, the dunes small or large, isolated or united
into a continuous ridge, the sand may be relatively quiet or in active
motion, the prevailing plant life may consist of various different
species, sometimes with almost no semblance of order or arrange-
ment. In some places the vegetation seems to have the upper
hand in the unceasing struggle with the wind, covering the top
and both sides of the dunes, and extending seaward down to the
actual limits of wave action. In others, the plants have difficulty


in maintaining themselves, or are actually yielding before the wind.
There great expanses of sand are almost or quite devoid of plants,
and sand blows continuously through gaps on the dune ridges and
piles up in great heaps on the leeward side, burying the shrubs
and small trees and threatening the houses and gardens which
are built too close.
But in this apparent chaos of topography, of vegetation, and of
wind action, a general correlation of vegetation with the environ-
ment may be discovered, which, though varying in detail from
place to place, is nevertheless uniform in its essential principles.
There are but four radically different types of environment, al-
though each of these varies considerably and most of them pass
into each other by gradual variations. The first type is found
immediately along the shore of the ocean within the limits of
wave action, where the influence of the salt water effectually
prevents all terrestrial plant life. In the second, the general
effect of the wind is to deposit sand. Such areas are found on
the lee side of every clump of plants, and, on a larger scale, on the
lee side of every dune opposite a break, where sand is pouring
through and accumulating in a heap. Here are found only such
plants as can withstand burying and are able to grow as rapidly
as the sand rises about them, or some trailing vines which maintain
their roothold at the side of the accumulating sand while their
long stems ride up with it. In the third, sand is being removed by
the wind. In such situations the development of seedlings is
extremely precarious and usually impossible, and the few plants
consist of old individuals established before excavation began.
In the fourth, the surface of the sand is stationary. Here numerous
species of plants grow, but the mass effect of all of them must be
to protect the surface so that the stability of the sand is maintained.
The apparent lack of order in the arrangement of plant life on
the dunes is due to the presence of intermediate variations be-
tween these typical environments. There is every degree of
gradation in the speed of sand removal, from places which are
nearly stationary to others in which no plants are able to grow,
and a similar complete gradation from areas of stationary level to
those where sand is being rapidly deposited. The number of
species and the number and size of individual plants reach their
maximum on the stationary sand, and decrease to the areas of
rapid removal or deposit. Nor is there any regularity in the space


arrangement of these four environments, except that the first
must necessarily lie adjacent to the water. The stable dune-
ridge may be broken by wind-sweeps through which the sand is
in continual motion, and the great flat beaches near the shore are
frequently marked by small dunes where plants are able to check
the movement. Some of these dunes are waging a losing struggle
and will soon disappear, while others are more successful and are
actually increasing in size.
It must be appreciated therefore, that there is no other place
in Porto Rico in which the surface of the land may change so

Fio. 3. Philoxerus vermicularis at the seaward margin of the dune vegeta-
tion; Sporobolus virginicus at the rear.
quickly, or in which the environment may alter so completely
from one year to the next. Under these extreme conditions, the
vegetation will necessarily be sparse, and limited to those species
which can endure the unusual stresses of wind and sand, and which
are able to take immediate advantage of every opportunity to
establish themselves in a new site.

A narrow or wide sandy beach parallels the shore of the Atlantic
Ocean throughout the extent of the Tertiary formations, except
where broken by projecting headlands of the San Juan formation
or by estuaries of the numerous rivers. On this beach the plant
life is essentially the same as that described in the previous sub-


series, except that more intense wave action, greater exposure to
wind, and a consequently less stable substratum serve to reduce
the number of plants to a minimum. Over long stretches of
beach no plants whatever may be found below the edge of the
dune thickets, and elsewhere their numbers are always small.
Philoxerus vermicularis (L.) Nutt. (Fig. 3) is usually rooted closer
to the ocean than any other species. It is prostrate, with red
fleshy stems spreading widely over the sand and somewhat as-
surgent at the tips. It contributes visibly to building up small
dunes and is conspicuous where wave action has begun to encroach
on the beach. Associated with it on the beach are Ipomoea Pes-
caprae (L.) Roth., rooted above the beach and trailing down upon
it, and Sporobolus virginicus (L.) Kunth, migrating into the beach
by shallow rhizomes. Scattered individuals of Cakile lanceolata
(Willd.) 0. E. Schulz, Diodia maritima Thonn., and Chamaesyce
buxifolia (Lam.) Small are rare.

Just above the limit of wave action, where the sand is relatively
quiet, several species of dune-forming plants appear. Among the
first of these are Chamaesyce buxifolia (Lam.) Small, Diodia mari-
tima Thonn., and Remirea maritima Aubl., all of which catch
the drifting sand and build up small dunes, which may reach a
width of one or two meters (3-6 ft.) and a height of one to three
decimeters (4-12 inches). These miniature dunes attract other
species, especially on the somewhat sheltered lee side, and of these
the most important is Coccolobis uvifera (L.) Jacq. This is the
dune-former par excellence of the region. Its roots extend widely
through the sand and its branching stems spread over the surface,
protecting the sand by its broad leaves from drying out and the
consequent wind action, and at the same time catching and holding
sand which is blown into the thicket. Under a cover of Coccolobis
the dune grows rapidly, soon unites with others near it, and the
whole eventually rises to the dignity of a continuous dune-ridge
(Pl. 5) parallel to the coast. On such a ridge, the stems of the
Coccolobis grow forward over the surface of the sand, occupy the
pioneer zone of the herbaceous species, and reduce them to a
few isolated colonies.
Some other species have the same general habit as Coccolobis
and occasionally form dunes in the same way (Pl. 6), but they are


less abundant in this habitat, form fewer dunes, and are in every
way less important. Among these may be noted Scaevola Plumierii
(L.) Vahl and Chrysobalanus Icaco L., both of which are far more
abundant and ecologically more important in other habitats, and
Opuntia Dillenii (Ker-Gawl.) Haw. The latter forms tangled
spreading mats becoming as much as 3 meters wide under favorable
circumstances. Under thickets of Coccolobis it grows slender and
erect, and is apparently soon shaded out by this dominant species.
After a dune has been established by a thicket of Coccolobis, it
may still continue to grow, but the maximum height probably
does not exceed 15 meters. With the attainment of maximum
height and stable conditions, a notable change in the environment
takes place. The shade of the Coccolobis thickets and the fallen
leaves on the ground shade the surface of the sand, help conserve
the soil moisture, and by their decay add to the organic content
of the sand. These conditions further the growth of various
other species, which appear in increasing numbers, both of species
and individuals, as the dune becomes older. Among the first to
appear are various species of herbs, which live in the shade of the
thickets, and of climbers, which trail over them. Among the
latter are Ipomoea dissecta (Jacq.) Pursh, Mikania congesta DC.,
Smilax coriacea Spreng., and Serjania polyphylla (L.) Radlk., while
the most characteristic herb is Hymenocallis declinata (Jacq.) M.
Roem. The introduced species Cordyline guineensis (L.) Britton and
Bryophyllum pinnatum (Lam.) Kurz are frequent and often form
large patches where the shrubs have been removed. Various weedy
species also come in as immigrants from the cultivated grounds
near the dunes. Certain other shrubby or arborescent species also
appear and mingle sparsely with the Coccolobis, such as Chrysobala-
nus Icaco L., Tabebuia pallida Miers, Calophyllum antillanum
Britton, Capparis flexuosa L., Plumiera alba L., and Pithecellobium
Unguis-cati (L.) Mart. None of these reaches a large size or
surpasses the general level of the Coccolobis, because of the exposure
to continuous winds which checks growth at a uniform level of
about one meter. Two species of more exposed situations, Ipomoea
Pes-caprae (L.) Roth. and Canavali maritima (Aubl.) Thou. trail
over occasional exposed places in the thickets and are there often
associated with Ipomoea stolonifera (Cyrill.) Poir.


With the first rise of the embryonic dune above the general level
of the beach, there is created a lee side with less exposure to the
wind, upon which sand is being gradually deposited. Here the
first colonizers are Diodia maritima Thonn., Remirea maritima
Aubl., Chamaesyce buxifolia (Lam.) Small, and Ipomoea stolonifera
(Cyrill.) Poir., while scattered trailing plants of Ipomoea Pes-
caprae (L.) Roth. and Canavali maritima (Aubl.) Thou. also
flourish. As the dune increases in size to the maximum and the

FIG. 4. Summit of a San Juan consolidated dune: the original dune vegetation
is reduced to a single wind-swept thicket of Coccolobis uvifera, surrounded by
a sod of Stenotaphrum secundatum.

deposit of sand on the lee slope ceases, this sparse vegetation
of the younger lee slopes increases into a dense thicket of shrubs
and small trees, in which Coccolobis uvifera (L.) Jacq. is the most
important, although in a few stations replaced almost entirely by
other species. On these stabilized lee slopes the conservation of
moisture and the addition of organic matter to the sand follows
just as on the ridge, and there is the additional advantage of pro-
tection against the constant winds from the ocean. The thickets
of the lee slope accordingly become higher, denser, and composed of
more species than those of the dune-ridge. Coccolobis uvifera,
which seldom exceeds a meter in height on the ridge, here grows


into a small tree, but is seldom or never able to rise above the level
of the summit of the ridge, beyond which height the wind prevents
any further growth (Fig. 4). The other associated shrubs become
similarly taller, while various other species occur, such as Cono-
carpus erecta L., Bucida Buceras L., Ichthyomethia piscipula (L.)
Hitchc., and Erithalis fruticosa L. Terminalia Catappa L. is
abundantly planted as a windbreak along the lee base of the dunes,
and numerous seedlings spread gradually up the lee slope. They
are most abundant where the original vegetation has been destroyed
but occasionally mingle to a small extent with Coccolobis and other
native species of the thickets. It also shows a pronounced sus-
ceptibility to wind; mature planted trees are regularly bent to one
side (PI. 7), with most of their branches on the south side, and the
young plants are unable to rise above the crest of the dune. The
various lianas of the dune-ridge still persist. Herbaceous vegeta-
tion is almost completely lacking under the denser thickets. Where
the shrubbery has been partly cleared for firewood, the herbs are
chiefly weedy species immigrating from the adjacent gardens and
The later developmental history of the dune thickets is not well
illustrated, since the young trees are freely cut for fuel and since
they are often disturbed or destroyed by breaks in the continuity of
the protecting ridge behind them. It seems obvious, however, that
additional species are slowly introduced by natural methods,
gradually breaking the dominance of the Coccolobis and its shrubby
associates and converting the thickets into a mixed forest. Among
the later species we have observed Elaphrium Simaruba (L.) Rose,
Zanthoxylum martinicense (Lam.) DC., and Ecastophyllum Ecasto-
phyllum (L.) Britton, with tangled masses of Guilandina Crista (L.)
Small, scattered vines of Stigmaphyllon tomentosum (Desf.) Ndz.,
and occasional plants of Zamia latifoliolata Preneloup. Most of
these are reminiscent of the second-growth thickets of the limestone
hills, and the stabilized dunes were doubtless originally covered
by a mesophytic forest very similar to that of the older limestone.
The history of a normal dune from its inception to its final stabili-
zation may be summarized as follows: As soon as the sand near the
beach is quiet enough to maintain plant life, it is occupied by pioneer
herbaceous species, which are soon displaced by thickets of Coc-.


colobis. Increase in the size of the dune leads to an improvement
in its environment on the crest and lee side, and this causes a steady
addition of species, so that the vegetation progresses from a very
few species, all adapted to a severe environment, to a large number,
adapted to a more normal environment and ultimately producing
a mixed forest. The windward slope of the dune .does not share in
this completely, since it remains permanently exposed to the full
force of the wind. There the thickets of Coccolobis persist and on
the shoreward margin of the vegetation, where environmental
conditions are always precarious, such plants as Sporobolus vir-
ginicus (L.) Kunth and Philoxerus vermicularis (L.) Nutt. still
form the outposts of the vegetation in position, just as they were
also the pioneers in time.

Not every little dune which appears along the beach carries out
this whole series of development. Many of them are destroyed
while young, either by being blown away completely by the wind,
because of an insufficient covering of vegetation for holding them,
or by the too rapid deposition of sand. A mature dune, completely
occupied by vegetation, is however permanent, unless the vegeta-
tive covering, to which it owes its continuous existence, is destroyed
by some catastrophe. Most of these catastrophic causes may be
referred to some human action. Even the establishment of a
well-worn path across the crest may lead to the removal of sand by
the wind, the undermining and destruction of the plants at its
side, and the gradual development of a great windsweep which
destroys the dune and produces a new area of active sand on its
lee side. Numerous instances of such destruction may be ob-
served along the beach, covering every stage from the first appear-
ance of erosion by the waves or the wind to the development of
large wind-sweeps with great expanses of bare sand.
But few cases may be found in which the destruction is being
caused by the action of the waves. Then the attack is against
the windward base of the dune, producing low but steep cliffs of
sand, held in place by the roots of the Coccolobis thickets and
favoring the growth of new plants of Philoxerus vermicularis (L.)
Nutt. Further development of such action to the complete des-
truction of the dune has not been observed.
In the great majority of cases, the destruction of the dune is


caused by a break in the vegetative cover on its crest (Pl. 8).
The sand is removed from this break and poured over to the lee
side of the dune, where it forms an actively growing lee slope.
At the same time, the break gradually widens, accompanied by the
undermining of the plants at its sides. This process produces
steep sides held by roots of the Coccolobis and its associates.
With the development of an outlet for the wind through the dunes,
more and more sand is removed from the windward side, carried
through the trough and added to the new lee slope, so that a bare
wind-sweep develops near the shore, and the general line of the
dunes is pushed farther back.
The fresh sand piled up at the lee of the dune is deposited on open
ground or on vegetation indiscriminately, and the usual dune thick-
ets are soon buried under the accumulation (Pl. 9). Some species
are able to withstand burying to a considerable extent, such as
Calophyllum antillanum Britton, and tend to keep above the dune
by prolonged growth. Serjania polyphylla (L.) Radlk. also has this
habit, and is frequently seen twining over partially buried thickets.
Canavali maritima (Aubl.) Thou. appears to be commoner here than
anywhere else, rooting mostly at the edge of the lee slope and
trailing over it. Cenchrus carolinianus Walt. and various other
herbaceous species occur more rarely.
In front of the break through the ridge, the sand is rapidly re-
moved, and the result is that a large flat expanse of beach (Pl. 10)
is produced, nearly or quite devoid of plants. Frequently on such
flats small dunes may be observed, the relics of larger ones of an
earlier stage in the dune activity, which have been nearly destroyed
by wind action. It is not always possible to differentiate between
an old dune of this type and a young dune of recent origin, but in
some cases this may be done. The old dunes usually are held by
old and large plants of some efficient sand-binder, like Coccolobis
uvifera (L.) Jacq., apparently larger than the size of the dune
requires, and have comparatively steep sides, showing the removal
of sand by wind action, while the young dunes are held by much
smaller plants and are relatively broad and low with gently sloping
sides. Coccolobis uvifera, the most abundant plant of the stabilized
dunes, is naturally the most abundant on the relic dunes also, but
other similar dunes may be held by any efficient plant which
happens to be established.' Borrichia arborescens (L.) DC.,
Scaevola Plumierii (L.) Vahl, Sporobolus virginicus (L.) Kunth, and


Opuntia Dillenii (Ker-Gawl.) Haw. were observed. Even though
such dunes are being gradually destroyed by erosion on the wind-
ward side, they still operate to collect sand in their lee, and that
area of deposition is at once occupied by Remirea maritime Aubl.,
Diodia maritima Thonn., Ipomoea Pes-caprae (L.) Roth., and the
other plants of the habitat. In one case a plant was noted of
Lantana involucrata L., a species most abundant on solid limestone.
All of these interruptions in the usual order of the developmental
history of the dune vegetation must be regarded as merely temp-
orary. Although they may cause a great destruction of vegetation

FIG 5. A break in the dune vegetation permits sand to blow through rapidly,
producing a steep windward slope and covering a cane railway.
and a considerable redistribution of sand and sometimes overrun
cultivated fields, roads (Fig. 5), and houses, they eventually
come to a close with the new stabilization of the sand. Sooner or
later the supply of sand is exhausted from the wind-sweeps along
the shore, while the loose sand farther inland is protected from the
wind by intervening ridges. Vegetation immediately recovers the
dune and it is soon returned to a condition of stability. At num-
erous points along the north shore compound series of dune ridges
show where this process of dune formation, destruction, and re-
capture by plant life has taken place in the past.


Under the densest thickets of Coccolobis on the active dunes,
the sand has already become comparatively firm and solid, and there
is reason to believe that the actual consolidation of the sand into
a solid limestone rock, by means of organic substances mixed with it,
begins rather promptly after stabilization of the dune by vegetation.
The completion of the process is, of course, a matter of long dura-
tion, probably to be reckoned by tens of thousands of years, and
the result is a limestone rock known as the San Juan formation.
It is entirely probable that the vegetation of the present extensive
ridges of San Juan limestone passed through the same general
stages of development, in prehistoric time, that now exist on the
modern active dunes, although the species of plants concerned
were possibly considerably different.
Since consolidation must have followed after stabilization of the
dunes, and since that process was dependent on vegetation, it is
apparent that the preliminary stages in the development of plant
life on the ancient San Juan dunes must have been completed before
their consolidation. Then the first vegetation on the limestone
rock must have been somewhat similar to that of the stabilized
dunes. The thin soil of the present San Juan limestone ridges is
produced by the disintegration of the limestone into a lime sand,
more or less altered by the leaching out of certain constituents and
by the addition of organic matter. This soil is essentially the
same as that now found in the stabilized dunes, which has a similar
origin, but with the addition of a thorough washing in sea water.
The vegetation of the two types of dunes is therefore very similar,
but the additional species on the San Juan dunes give evidence of
the greater and longer continued stabilization of the soil and the
greater accumulation of humus.
These agencies, which have made some effect even on the vegeta-
tion of the crests and sides of the dunes, reached their maximum
efficiency in the valleys between, where erosion from the slopes has
created a deep and fertile soil. Certainly the natural vegetation
there was of a different type, but no examples of it have been seen,
the land being all converted to agriculture. Even the sides and
crests of the dunes are generally cultivated, although two localities
were observed in which the summits of the dunes were characterized
by a very thin soil and much rock exposure, without value for agri-


culture. These two summits were still occupied by a remnant
of the original vegetation. They have certainly been cut over re-
peatedly for firewood and building material, so that the natural
vegetation has been changed by the gradual elimination of some
of the more valuable species of trees, and the greater development
of some other species.
The present vegetation is a dense thicket of shrubs or young
trees, in which Chrysobalanus Icaco L. is by far the most abundant,
forming almost impenetrable thickets around the margin and being
common through all parts of the tract. A few plants of Coccolobis
uvifera (L.) Jacq. still persist in the more exposed places. Num-
erous other species found likewise on the active dunes still show the
similarity in environmental conditions, but many additional
species also occur. Among the commoner plants are Calophyllum
antillanum Britton, Capparis flexuosa L., Tabebuia pallida Miers,
Distictis lactiflora (Vahl) DC., Calyptranthes Sintenisii Kiaersk.,
Gouania lupuloides (L.) Urban, Serjania polyphylla (L.) Radlk.,
and Trichostigma octandrum (L.) H. Walter, with numerous weedy
species around the margin, such as Solanum persicifolium Dunal,
Borreria verticillata (L.) Meyer, Rivina humilis L., Jacquemontia
nodiflora (Desv.) G. Don, Bromelia Pinguin L., and Bryophyllum
pinnatum (Lam.) Kurz. On the thin ledges of rock in the shade
are Pilea microphylla (L.) Liebm., Clavenna tetrandra (L.) Standl.,
and various other small herbs. On the summit rise the tall trunks
of Elaphrium Simaruba (L.) Rose and on the shadier north side
such mesophytic trees as Ficus laevigata Vahl and Clusia rosea Jacq.
occur, while tall climbers of Philodendron Krebsii Schott and
Vanilla Eggersii Rolfe are abundant, and the exposed ledges are
occupied by large clumps of Anthurium acaule (Jacq.) Schott.
Hylocereus trigonus (Haw.) Safford is common in the trees.
The similarity between these thickets and those of the limestone
hills in general appearance, floristic composition, and environment
is striking, and leaves little room for doubt, notwithstanding their
present fragmentary development, that they represent the same
plant association. We may accordingly assume that the original
mesophytic forest of the limestone hills, now completely destroyed,
also extended over the consolidated San Juan dunes, and repre-
sented the temporary climax of both mesarch and xerarch succes-
sional series.


It has already been stated that two, or possibly three, series of
the consolidated dunes of the San Juan series were formed. Due
to comparatively recent subsidence of the land, the second series
is now in contact with the sea and partly below its level (Pl. 11)
and subject to the disintegrating work of the waves, which are
rapidly reducing the rock to lime sand. This mechanical action
is followed closely by retrogressive successions in the vegetation.
Along the lee front of the outer ridge, pasturage and clearing,
aided by the effect of a salt-laden atmosphere, have reduced the
original forests to mere thickets, composed largely of thorny
species. In such localities the vegetation now consists of dense
mats of Opuntia Dillenii (Ker.-Gawl.) Haw., crowded bushes of
Lantana involucrata L., tangled masses of Guilandina Crista (L.)
Small, and thickets of Coccolobis uvifera (L.) Jacq. Other common
shrubby species include Elaeodendrum xylocarpum (Vent.) DC.,
Plumiera alba L., Anthacanthus spinosus (Jacq.) Nees, Argythamnia
candicans Sw., Gundlachia corymbosa (Urban) Britton, Chrysobalan-
us Icaco L., and Ichthyomethia piscipula (L.) Hitchc. They are
aggregated into dense and almost impenetrable thickets, seldom
more than one or two meters high, and tangled with vines of
Stigmaphyllon tomentosum (Desv.) Ndz., Distictis lactiflora (Vahl)
DC., and Serjania polyphylla (L.) Radlk. The small open places
between the thickets are carpeted with a thin sod of Stenotaphrum
secundatum (Walt.) Kuntze and also colonized by numerous weedy
Toward the top of the ridge, where there is-full exposure to the
wind, many of these species disappear. Three of the four com-
monest species of the lee slope are still dominant, Guilandina Crista
(L.) Small having become greatly reduced or completely lacking.
Gundlachia corymbosa (Urban) Britton becomes more abundant,
Eugenia axillaris (Sw.) Willd. appears, and one colony of Hymeno-
callis declinata (Jacq.) M. Roem. was noted. The summits of these
ridges, when the bases are fully exposed to the sea, frequently
show longitudinal fissures, indicating lines along which blocks of
stone will be broken off.
On the outer or windward slope of the ridges the action of wind
and wave is at the maximum. The direct impact of the waves
causes undercutting, changing the side of the ridge from a gentle


slope to a steep or vertical cliff (Fig. 6). This later becomes
overhanging and then large blocks drop off into the ocean to be
ground up by the waves. Caverns are frequently excavated at
or near sea level and fissures opening into them from above are
widened by erosion into large pits. It is by no means uncommon
to see spray from the waves spout high above the dunes like a
geyser from some opening invisible from the side. If the ridge is
not too high, spray and waves dash over them to their summit.
There the water removes all surface soil, erodes deep pockets and
fissures in the limestone, and carries away the rock by erosion and

FIG. 6. Seaward face of a San Juan consolidated dune west of Arecibo.
Undercutting of the cliff and disintegration of the limestone by wave action is
in active progress.

The erosion of the comparatively friable limestone by the waves
produces a most interesting series of effects, due partly to solution
and partly to physical action. The whole surface of the limestone
is honeycombed into countless pockets and ravines, of every con-
ceivable width and depth, separated by sharp and jagged edges.
The pockets, when filled with water, drain over the lowest part of
their margin into the next pocket below, but retain in their bottom
any loose sand which may have been washed in from above. Those
which lie in the usual path of escaping salt water soon become
permanently drained by the erosion of a deeper outlet, and all the


sand is washed away and carried down to the ocean, while still
others drain through sinkholes.
The general effect of the action on plant life is irresistible destruc-
tion, in which the various species of the limestone ridge disappear
gradually, leaving only the hardiest to carry on the struggle until
they in turn also disappear before the ceaseless action of the waves.
On the higher ridges, where the sea water does not dash over the
crest, the thickets persist on the landward side, and thin out at the
top as mentioned above. Down the seaward side, even the hardiest
eventually drop out, as the exposure to salt spray and soil erosion
becomes more complete.

FIG. 7. Low thickets of Coccolobis uvifera on the San Juan consolidated dunes.

In the flat deposits of sand in the largest pockets, provided they
stand so high that they are seldom filled by salt water and probably
depend more on rain water for their origin and filling with sand,
Ipomoea Pes-caprae (L.) Roth., Canavali maritima (Aubl.) Thou.,
and Sporobolus virginicus (L.) Kunth persist, gradually giving way
at lower levels where the salt water fills the pockets frequently.
Seedlings of Sporobolus virginicus and dwarf plants of Erigeron
bellioides DC. appear even in the small pockets and at lower eleva-
tions but are quickly destroyed by salt water. Coccolobis uvifera
(L.) Jacq. (Fig. 7), Chrysobalanus Icaco L., Scaevola Plumierii
(L.) Vahl, and Borrichia arborescens (L.) DC., cling tenaciously to
small fissures on the side of vertical walls, becoming smaller in


size and fewer in number at the lower elevations, where sea water
strikes them more frequently, and eventually clinging only to
crevices on the lee side of the walls. Borrichia arborescens (L.)DC.
seems to extend farther toward the ocean than any other species.
At numerous places along the north shore breaks in the continuity
of the San Juan dunes may be observed. These may be only a few
meters wide and supposedly of very recent origin, and range up to
a kilometer or more wide and probably very old. In every case
the penetration of sea water to the inner face of the San Juan
ridge leads to the development of a roughly semicircular bay (Pl.
11) opposite the break. Sand produced by the disintegration of
rock on the outer side of the ridge is carried westerly along the
shore and much of it is washed inland and deposited around these
bays. At the same time, the breaks serve as wind-gaps, and cause
the development of great wind-sweeps between the San Juan ridge
and the accumulation of the present active dunes. In such places
the wind, and in some places the waves also, attack the vegetation
of the ridge from the inner or lee face (Pl. 12). On the lower
ridges also, now nearly or quite surrounded by the ocean, the spray
and salt water dash over the summit and help destroy the vegeta-
tion of the lee side. Under these circumstances Coccolobis uvifera
(L.) Jacq. (Fig. 7) and Borrichia arborescens (L.) DC. appear to
be the hardiest species, clinging to thin deposits of sand and some-
times forming extensive thickets in which the tallest plants are
not more than 3 decimeters high. But even here, each fresh
deposit of a thin layer of sand behind a rock is at once occupied by
pioneer plants of Sporobolus virginicus (L.) Kunth. Near the
level of the sea on the lee side of the ridges is a favorite habitat for
Suriana maritima L.
The whole process is one of irresistible and unceasing wave
action, steadily destroying the rock and decreasing the amount of
vegetation more and more until it is eventually destroyed. This
action is continuous below, near the ocean, but is intermittent at
higher levels, and there each brief respite is utilized by the vegeta-
tion to attempt another advance toward the sea.

Wave action deposits sand on the beach, which is blown inland
by the wind. Xerophytic plants, initiating a xerarch successional
series, collect and stabilize the moving sand into dunes, charact-


erized by Coccolobis uvifera (L.) Jacq. and associated species.
Favored by the reduction in exposure to wind and by the gradual
addition of organic matter to the sand, numerous other species
immigrate and produce a dune thicket. Renewed exposure to
wind caused by breaks in the -continuity of the protecting dunes
lead to the destruction of the thickets and the reversion of the
vegetation to its pioneer stage. In the course of time the sand
of the stabilized dunes, if they are not previously destroyed, is
consolidated into limestone and occupied by a mesophytic forest
essentially identical with that of the limestone hills.
Submergence of the land exposes the limestone to the waves which
gradually disintegrate it into lime sand, and also exposes the
vegetation to strong winds and a salt-laden atmosphere. The
forests become reduced to mere thickets and eventually destroyed.
The sand produced by the disintegration of the limestone is washed
and blown inland and contributes to the. formation of new dunes,
initiating a new cycle of'theame type.
The general tendency of,4the vegetation is toward the develop-
ment of a mesophytic forest, and this trend will be hastened by
elevation of the coast and retarded by its further submergence.

C. The Vegetation of Icacos Cay '
leacos .Cay is a small island lying a few kilometers offshore
from-iFajardo., Playa. It is conrposed of a core of limestone, partly
surrounded by maritime deposits of beach sand, and in its vegeta-
tion, which is quite unlike that of the mainland, i- probably typical
of numerous other similar islands in the same region. Its flora
has already been listed by Britton (9).
The limestone of which the island is composed is solid and very
white, and rises in flat-topped, irregular hills to a height of about
30 meters (100 ft.) above the sea. The hills are not continuous,
but include at least three undrained depressions and several small
sinkholes. The rainfall is scanty, probably not exceeding 100 cm.
(40 inches) per year, and surface drainage is reduced to a minimum.
The surface of the limestone is greatly broken, pitted, and eroded.
In many places the only available soil is in the pits and crevices
and elsewhere it is very thin and stony. The western end of the
island is composed of a flat shelf, lying about two meters (6 ft.)
above sea-level, underlaid by limestone and covered with a thin
layer of sand. A sandy beach is well developed around the leeward


side of the island, but on the windward side the limestone descends
to the water's edge, and littoral deposits are mostly limited to
small quantities of sand on rock ledges and in crevices.
The sandy beaches are occupied by a zone of Coccolobis uvifera
(L.) Jacq. and the species usually associated with it. The flat
land behind the beaches has been extensively cleared and partly
planted to coconuts, so that the original vegetation is destroyed.
It was probably closely similar to the beach thickets described
above (p. 50), or, in its older portions, may have progressed into
a xerophytic forest. Along the windward side of the island the Coc-

FIG. 8. Salt pond on Icacos Cay, surrounded by an interrupted fringe of
Rhizophora Mangle.

colobis thickets are reduced to a narrow strip at the foot of the
limestone hills and there lie in non-successional contact with the
xerophytic vegetation of the hills.
The largest undrained depression is occupied by a pond (Fig.
8) of salt water, surrounded by a narrow and interrupted zone of
mangroves, in which Rhizophora Mangle L. is prominent. Two
other depressions lie far above the influence of salt water, but are
still occupied by a few halophytic species. In one of them a
central open space is colonized densely by Batis maritima L., sur-
rounded by a zone of Conocarpus erecta L.; in the other Conocarpus
has filled the entire depression. In each case, their position favors
the gradual leaching out of salt by percolation, while the soil will


be kept moist by drainage from the surrounding hills. This
should favor the establishment, in time, of a vegetation somewhat
similar to that of the limestone hills described above, (p. 39) and
an indication of this development is already seen in the presence
of Ficus laevigata Vahl.
The limestone hills are covered with a great expanse of xero-
phytic chaparral (Fig. 9), seldom exceeding one or two meters
(3-6 ft.) in height. Lantana involucrata L. is everywhere the
dominant species, and is associated with smaller numbers of
other microphyllous shrubs. Among these Jacquinia Berterii

FIG. 9. Icacos Cay: the limestone plateau is covered by a xerophytic chaparral,
in which Lantana involucrata is the dominant species.

Spreng., Pisonia subcordata Sw., Randia mitis L., Corchorus hirsutus
L., Croton betulinus Vahl, Rhacoma Crossopetalum L., Argythamnia
candicans Sw., Plumiera alba L., Pithecellobium Unguis-cati (L.)
Mart., Wedelia calycina L. C. Rich., Bumelia obovata (Lam.) A.
DC., Chamaesyce articulata (Aubl.) Britton, Anthacanthus spinosus
(Jacq.) Nees, and Jacquinia Barbasco (Loefl.) Mez are common.
Tall specimens of Cephalocereus Royeni (L.) Britton & Rose, and
Leptocereus quadricostatus '(Bello) Britton & Rose, with mats of
Opuntia Dillenii (Ker.-Gawl.) Haw. are conspicuous but not
abundant. There are a few lianas of Jacquemontia jamaicensis
(Jacq.) Hall. f. and Urechtites lutea (L.) Britton, but herbaceous
vegetation is almost wanting.


Along the summit of the limestone hills and fully exposed to the
trade winds is a dense thicket of wind-swept one-sided shrubs of
Capparis cynophallophora L., Capparis flexuosa L., Elaeodendrum
xylocarpum (Vent.) DC., Pisonia subcordata Sw. (Fig. 10), Jac-
quinia Barbasco (Loefl.) Mez, and Strumpfia maritima Jacq.
(Fig. 11).

FIG. 10. Icacos Cay: Pisonia subcordata is abundant along the crest of the
ridge. The shrubbery is wind-shorn and the plants one-sided.

The numerous rivers which empty into the ocean on the north
shore of Porto Rico are all comparatively small and subject to
great fluctuations in the amount of water. They follow tortuous
courses through the mountains and the limestone hills at the bottom
of narrow valleys. Immediately after passing the last belt of hills.


where they descend nearly or quite to sea level, they broaden out
into estuaries and the swift current of the upper course is greatly
reduced, causing the river to deposit its burden of eroded materials.
As a result of this action, continued for long periods of time and
sometimes favored by changes in the general level of the coast
line, the rivers have built up extensive areas of alluvial soil, which
are often termed playa lands. In the soil survey of the Arecibo
to Ponce Area, the soil of these tracts is named Arecibo silt loam.
At present these delta lands are seldom overflowed, and there is
reason to believe that most of them were formed as submarine

FIG. 11. Icacos Cay, wind-swept summit of the central ridge. The vegeta-
tion is xerophytic with several halophytic species. The single tall shrub is
Strumpfia maritima.

littoral deposits. Under such an origin, it is reasonable to presume
that their first vegetation after uplift above the sea was a halophytic
vegetation of mangroves, and that continued elevation and the
leaching out of their salt content gradually carried them through
a successional series with unknown intermediate stages into a
climax mesophytic forest.
This forest is at present completely destroyed. Mr. E. D.
Col6n, who has seen some of the last remnants of it near Barceloneta
informs us that it contained, among other species, maria (Calophyl-
lum antillanum Britton) and a roble (Tabebuia sp., possibly T.


pallida Miers). In that location, isolated small trees of these
two species still exist as proof that both of them can grow under
this environment.
At the present stage of elevation of the land, the rivers are
usually intrenched throughout their lower courses in deep channels
with vertical walls of mud, and successions to a forest are apparently
not in progress, due to complete utilization.of the land for agricul-
ture. Quiet waters along the banks are often occupied by small
colonies of Castalia ampla Salisb. and Piaropus crassipes (Mart.)
Britton. At the edge of the stream is a fringe of amphibious
vegetation, such as Sagittaria lancifolia L. and Typha angusti-
folia L. Thickets of the tall grasses Phragmites Phragmites (L.)
Karst. and Gynerium sagittatum (Aubl.) Beauv. are conspicuous
but scattered features of the plant life.
The playa lands are excellently adapted to cane cultivation and
are planted almost exclusively to that crop.

At various places along the north shore there are extensive
developments of fresh-water lagoons and swamps, produced by
various physiographic causes. Where the playa lands slope gently
down to sea level and are separated from the shore by a ridge of
sand dunes, the natural drainage is obstructed and the water tends
to accumulate as small swamps. In other places it appears that
the subsidence of the land has drowned the mouths of certain rivers
and caused their water to escape from the channels into the low
land behind the dune ridges. The largest fresh-water lake is
Laguna Tortuguero (Pl. 13), northeast of Manati, about 4 kilo-
meters (2.5 miles) long and one-half kilometer (one-fourth mile)
In the quiet water of the lagoons and in the small drainage
ditches frequently cut through the smaller swamps, water of proper
depth is occupied by dense colonies of Castalia ampla Salisb.
Nearer the shore and less abundant are similar but smaller colonies
of Nymphoides Humboldtianum (HBK.) Kuntze, while Pistia
Stratiotes L. and Piaropus crassipes (Mart.) Britton also occur.
Submerged masses of Ceratophyllum demersum L. are common, and
the floating leaves of Potamogeton fluitans Roth are occasionally
Along the shore, and extending out to a depth of nearly a meter,


Typha angustifolia L. forms huge colonies, frequently acres in
extent. It is sometimes mixed with Mariscusjamaicensis (Crantz)
Britton, but the latter is more abundant and forms pure colonies
just at the shore line. Almost no other plants are associated
with these two tall dominant species, which have nearly a complete
monopoly of the habitat, but a few plants of the aquatic fern
Blechnum indicum Burm. establish themselves here and there
above water-level in the clumps of Typha.
Typha and Mariscus are both rapid soil-formers, producing a
black mucky soil by the decomposition of their fallen stems and
leaves, which gradually raises the ground above water-level. They
are so tenacious of their hold on the land that they follow the
change in level for some distance above the water, and at their
landward edge probably came in direct contact with an association
of shrubs or with the mesophytic forest of the playa lands. We
did not observe any such contacts, however, since the playa lands
are all under cultivation and the cane fields have been extended as
far into the lagoons as the cane can grow.

Where the Typha and Mariscus plants have been destroyed at
the edge of the marshes, uncovering a flat area of black muck, a
secondary association of several mud-loving species (Pl. 14) has
been developed. Here the soil is always saturated, numerous
small pools of stagnant water persist, and the water-table is never
more than a few centimeters below the surface. The areas of this
association are always relatively new and dominance by a particular
species or group of species has not been established. The associa-
tion is also short-lived, since both Typha and Mariscus tend to
re-establish themselves over the mudbanks as rapidly as their means
of propagation will permit.
The largest plants of the mudbanks are various species of slender
erect sedges, growing in dense colonies with a decided tendency to
monopolize the space. The most abundant of them is Eleocharis
interstincta (Vahl) R. & S. Here also are found Fimbristylis
complanata (Retz.) Link, Rynchospora cyperoides (Sw.) Mart.,
Eleocharis caribaea (Rottb.) Blake, and a few other species of
similar habit. Between the clumps of sedges, and over all the
surface of the mud before the sedges are established, are broad
mats of the creeping Bramia Monnieri (L.) Drake, usually con-


spicuously covered with bluish-white flowers, and more rarely
similar mats of Hydrocotyle verticillata Thunb. and Centella asiatica
(L.) Urban. Scattered plants of a few taller aquatic species, such
as Sagittaria lancifolia L., also occur.

The south shore of Laguna Tortuguero is immediately in contact
with the white sands described above. The white sand thickets
stand a meter or two above the water; at their margin there is an
abrupt slope downward to the nearly flat beach. The latter
consists of alternating areas of variable extent of mud-bank and
wet sandy beach which differ greatly in their plant life. The
mud-bank is related successionally to the marsh vegetation of
the lagoon, but the sandy beaches are sparsely occupied by a
group of species which represent merely an interpolation caused by
immigration, between two types of plant life which have no suc-
cessional relations whatever.
In the open sun in such locations, there is a scattered covering
of clumps of Scleria hirtella Sw., Fimbristylis diphylla (Retz.)
Vahl, and Aeschynomene sensitive Sw., with occasional plants of
various weedy species, such as Vernonia cinerea (L.) Less. In
the shade of the overhanging thickets, the vegetation is even
sparser, and the commoner species include Xyris Jupicai L. C.
Rich., Setiscapella subulata (L.) Barnh., Setiscapella pusilla (Vahl)
Barnh., Ibidium tortile (Sw.) House, Sauvagesia erecta L., and
Acisanthera Acisanthera (L.) Britton.

The thickets which lie between the wet beach and the dry
white sands at Laguna Tortuguero present an interesting assem-
blage of species. The water-table here lies well below the surface,
but the sand is kept constantly moist by capillarity, so that the
environment is distinctly mesophytic. The shrubs which compose
the thickets are from 1 to 5 meters in height, and many of them
would soon become trees if they were not repeatedly cut for fuel.
Chrysobalanus Icaco L. is the commonest species, and prefers the
drier or more exposed situations. With it and next in abundance
are Miconia prasina (Sw.) DC. and Miconia racemosa (Aubl.)
DC. Above the mud banks Ecastophyllum Ecastophyllum (L.)
Britton and Annona glabra L. are common. Among the other


species, some of the more abundant are Guettarda scabra (L.) Lam.,
Tournefortia hirsutissima L., Eugenia monticola (Sw.) DC., Icacorea
guadelupensis (Duch.) Britton, and Randia mitis L. Clidemia
hirta (L.) D. Don is the most abundant herb, and ferns are repre-
sented by Odontosoria aculeata (L.) J. Smith.
Part of the plants composing these thickets undoubtedly repre-
sent invaders from the semi-xerophytic habitat of the adjacent
white-sand thickets, and others probably indicate a tendency
toward a succession by the now extinct lowland forest.

The hydrarch series of associations is correlated with the usual
physiographic process of soil formation, and is represented by a
series of associations differentiated largely by depth of water or of
the water-table, and partly by the nature of the soil. Deeper
water is occupied by submerged or floating plants, shallower by a
marsh vegetation of Typha and Mariscus, and the land permanently
above water by a mesophytic forest. Temporary associations
follow the disappearance of the marsh plants on exposed muddy
beaches, or appear between the marshes and the adjacent forest
of another successional series. The climax vegetation is now
completely destroyed.

There is every gradation from seashores with considerable wave
action, where sand or even shingle can be transported by the waves
and accumulated on shore as dunes or shingle beaches, to others
protected from the prevailing winds and therefore free from effective
wave action. Most of the north shore of Porto Rico is of the
former type, and sand dunes, sand ridges, and sand beaches are of
common occurrence. There are, nevertheless, many places of more
restricted size where wave action is at a minimum. Two general
kinds of these shores may be noted. The first includes the banks
of river estuaries, where salt water is driven inland by the tidal
flow and produces saline conditions for some distance upstream.
In the second, which is by far the more important in Porto Rico,
the sea has broken through the littoral defense of dunes and over-
flowed the low swampy lands at the rear. These overflowed areas
become salt-water swamps, lagoons, or bays, depending on the
depth of water and the extent covered. In the largest, such as San


Juan harbor and the great lagoons east of San Juan, the depth may
be considerable, but the wave action is still much less than on
exposed shores. These habitats initiate a series of successions,
which proceed to a climax through intermediate stages charact-
erized by progressively decreasing salinity.
Most, but not all, of the swamps along the north shore are of this
type and show a fundamental similarity in their vegetation, but
the general agreement in the plant life is subject to much variation
caused by differences in the area, depth of water, salinity, and age
of the swamps. The larger present broad expanses of a uniform
vegetation extending over hundreds of acres, while in the smaller
the area of each plant association is limited. The deeper show
considerable areas of open water, with the usual halophytic vegeta-
tion confined to relatively narrow marginal strips. In those with
the most direct connection with the sea and consequently with the
most pronounced saline environment, the vegetation is composed
largely of distinctly halophytic species; in those with relatively
fresh water, the halophytes live only at the immediate shore,
and are replaced by associations of the hydrarch series at a short
distance back from the water. In general, the nature of the vegeta-
tion is determined by the depth and salinity of the water, while
the area of each association depends on the gradient of the ground
surface, a gradual gradient producing uniform conditions over
a large area, with similarly extensive development of each associa-
tion, and a sharper gradient restricting the various associations to
narrow zones parallel to the margin. Three well-marked associa-
tions of this series exist in Porto Rico, and it will also be necessary
to discuss the relations between these associations and others of
the xerarch and hydrarch series.

Throughout the tropics, the typical pioneer vegetation of such
shores with quiet salt water is a swamp forest, known as a mangrove
swamp or a manglare, composed of an aggregation of arborescent
species collectively known as mangroves or mangles, and represents
the initial stage of the series. Many different species of mangroves
exist, belonging not only to different genera but even to widely
different families of plants, and the similarity of name is due not
to their actual relationship but to their close agreement in many
features of structure and behavior. Mangroves constitute a dis-


tinct ecological unit rather than a taxonomic group, and most
of the features which they possess in common are in some way
correlated with their environment. The most remarkable of these
features is vivipary. In most species, irrespective of their taxo-
nomic affiliations, the seeds germinate before falling from the parent
tree. Since they remain in organic contact with the parent, and
derive a continuous supply of nourishment from it, the seedlings
are able to reach a large size and, when they are finally detached,
they are either planted in the mud directly by the impact of their
fall or they may drift in the water and be washed ashore and
anchored at some distance from the parent. Of the Porto Rican
species, Rhizophora Mangle L. illustrates this habit particularly
well. After the seeds germinate, the cotyledons enlarge relatively
little, but the hypocotyl continues to grow into a long fusiform
organ eventually 2-3 decimeters (8-12 inches) or even more in
length and 2-3 centimeters (1 inch) thick. They may remain
attached to the parent tree for a year, and become very conspicuous
as their size increases. Falling eventually from the tree, they
are easily anchored in the mud and are at once in a position for
growth, but they endure immersion in salt water for some time,
may be carried to some distance by currents, and serve as an
efficient means for the migration of the species.
Their second prominent adaptational feature is the development
of specialized roots. In Rhizophora Mangle L., these take the
form of prop roots. They arise from the branches, sometimes as
much as 5 meters above the water, grow down vertically or at an
angle and anchor in the soil when they reach it. In some cases they
branch before reaching the ground. Many such roots are pro-
duced on a single tree, forming with their branches an interlacing
tangle of roots and converting a colony of trees into an impenetrable
thicket. In Avicennia nitida Jacq. and many other species, the
specialized roots arise beneath the surface of the water and grow
vertically upwards into the air. Such roots are known as pneu-
matophores, and are regarded as functioning in the transportation
of oxygen from the air to the submerged root system.
Mangrove leaves, also, have a fairly constant structure and, ex-
ternally, a similar shape and appearance. In the Porto Rican
species, they are all elliptical in shape, or nearly so, thick in texture,
smooth and dark green in surface, and usually stand in a more or
less vertical position. All of these features seem to be correlated


in one way or another with the physiological effects of their peculiar
environment and have been made the subject of frequent investiga-
Mangroves are efficient builders of land. Their crowded stems,
aerial roots, and pneumatophores retard greatly or stop completely
the movement of sea water, both as waves and as currents. Any
load carried by the water in suspension is dropped among the trees,
gradually raising the land to the level of high tide. Since the man-
groves regularly grow in quiet water, the suspended materials are
the finer particles of silt and clay, rather than sand, and produce
a deposit of mud. To this is added annually the fallen leaves and
twigs of the mangrove trees, which by their decay add organic
matter to the soil and produce a black muck. As soil is accumu-
lated, the zone of mangroves pushes steadily seaward into deeper
water, extending the area of the land, but the rate of progress
necessarily becomes slower as deeper water is reached.
Four species of mangrove are regularly associated with the man-
grove swamps of Porto Rico (Pl. 15), Rhizophora Mangle L.,
known as mangle sapotera, Laguncularia racemosa (L.) Gaertn. f.,
the mangle blanco or mangle bobo, Avicennia nitida Jacq., the
mangle negro, and Conocarpus erecta L., the mangle botoncillo.
Of these the first three are by far the most important, and are
strictly limited in their growth to mangrove swamps, while the
fourth is usually smaller in size, relatively unimportant in the
swamps, and more abundant in other habitats. They differ also
to some extent in their salt requirement. Rhizophora Mangle L.
demands the highest concentration of salt and is not found far
inland from the ocean. Laguncularia racemosa (L.) Gaertn. f. and
Avicennia nitida Jacq. are associated with it along the seaward
margin of the swamps, but toward the landward side become more
common and finally displace it completely. In this part of a
swamp Concocarpus erecta L. usually appears and extends inland
into habitats which are nearly or quite free from saline conditions.
The first three, also, require an abundance of free water and flourish
only in a saturated soil, while Conocarpus withstands drier condi-
tions. It is accordingly found more abundantly in drained man-
grove swamps, or on the slightly higher land bordering the landward
margin of the swamps, and also occurs on sand dunes and even on
limestone rocks near the sea.
Under natural conditions, the mangrove swamps consist of pure


stands of the first three or all four of these trees, without additional
secondary species. The presence of additional species probably
always indicates a reduction in the salinity of the water. Malache
scabra B. Vogel becomes 2 meters (6 ft.) high and forms dense
thickets in the water of the narrow canals (Fig. 12) which traverse
the swamps. Pariti tiliaceum (L.) St. Hi]. is a widely branching
shrub or small tree, as much as 5 meters (15 ft.) high, growing in
the occasional open places among the mangroves, or extending out
over the canals. The stiff, thorny, sparsely branched stems of

FIG. 12. Canal through the mangrove swamp, Caflo Tiburones: at the right,
thickets of Malache scabra; at left, ascending stems of Drepanocarpus lunatus.

Drepanocarpus lunatus (L. f.) G. F. W. Meyer become 5 meters
(15 ft.) high, and arch out into the open sunlight above the canals.
Mangrove swamps are an important source of excellent fuel, and
their high economic value is fully recognized. Occupying land at
sea level and therefore incapable of drainage, except by an ex-
pensive system of dikes and pumps, they return a regular income
from hundreds of acres which would otherwise be useless. After
cutting they re-establish themselves easily and quickly.
As the mangroves build the land seaward, the landward margin
is gradually kept farther from the shore. The dense thickets
interfere with currents and tides and reduce the penetration of
fresh salt water. At the same time, rains and fresh-water streams


tend to dilute the salt-content at the shoreward margin. As a
result of their own activities, therefore, the environment of the
shoreward mangroves soon departs from the optimum conditions
for the association, and becomes suitable for a different type of
vegetation, resulting in succession.

A familiar type of vegetation along the north shore, well illus-
trated on the highways leading out of San Juan, is the association
of tall ferns, Acrostichum aureum L. It forms long strips or more
commonly irregular patches (Pl. 16), frequently several acres in
extent, along the landward margin of the mangrove swamps ofr on
slightly higher land within them. The soil is the usual black muck
of the mangrove swamp, frequently somewhat drier because of full
exposure to wind and sun, and always less impregnated with salt.
Over these areas, Acrostichum aureum L. grows in crowded masses
with fronds a meter (3 ft.) or more high, and forms an almost
pure stand, the secondary species being either completely missing
or so few in number as to be of virtually no ecological importance.
The position of the Acroslichum association in the successional
series is not entirely clear. Its great abundance and uniform
development in so many different localities and its regular position
immediately in contact with the mangroves is slightly drier or less
saline situations lead at once to the conclusion that it is the associa-
tion which normally follows the mangroves. On the contrary
there are some reasons, based on general ecological experience or
on actual conditions in Porto Rico, for regarding the Acrostichum
association as a temporary or secondary development.
It may usually be observed that the best development of the
fern association is near mangrove swamps which have been cleared
for fuel and in which the height of the trees is seldom more than
5 meters. Some of the fern associations also show clearly the
stumps of mangroves, which have recently been cut off and for
some reason failed to re-establish themselves. The succession of
a forest type of vegetation by a herbaceous type, except as the
result of some catastrophe, is unusual, to say the least, and certainly
not to be expected in Porto Rico, where every climatic condition
along the north shore favors the development of forest. Further-
more, it may frequently be observed that numerous young man-
groves are appearing around the margin of the Acrostichum associa-


tion, where they, as well as the very sharp transition line, indicate
that the actual succession is at present proceeding in the other
Acrostichum aureum L. (Fig. 13) is physiologically adapted to
living in the salt or brackish soils of the mangrove swamps, to
which few other herbaceous species, and none of similar size, are
adapted. Its microscopic spores are produced in enormous
quantities, and are doubtless scattered far and wide over all the
coastal plain every year, but in the mangrove swamps only a few
spores, coming to rest in a situation with better light and out of
direct contact with salt water, are ever able to grow. Under
normal conditions, it is represented by widely scattered individuals
and does not form an association of its own. But if the mangrove
trees are destroyed from a tract slightly above sea level, and that
is precisely the place where they are most subject to destruction,
because of the ease of working conditions, the place is at once
sowed with fern spores and soon occupied by the dense stand of
mature plants. Because of its density, the general invasion of
seedlings of tree species is retarded, but around the edge, where the
ferns can be gradually shaded out by overhanging branches, the
mangrove forest slowly advances and ultimately reoccupies the
whole area. Its position is therefore somewhat analogous to that
of Pteridium aquilinum (L.) Kuhn, also known as bracken fern,
in the northern pine forests, which becomes extraordinarily abund-
ant after clearing.

Not far from the shore of Humacao Playa is a forest tract of
large size in which the dominant tree species is Pterocarpus officinalis
Jacq. While a similar forest is not known elsewhere in Porto
Rico, and although this one does not come in direct contact with
typical mangrove swamps, we regard it as representing the next
stage in the halarch series.
The small rivers which flow down to the sea at this place reach
sea-level at a considerable distance from the present shore. The
small bays into which they originally emptied have been filled with
black soil, almost certainly through the activity of mangrove
swamps, and the shore line has been straightened. As the develop-
ment of these alluvial lands approached the open sea, exposure to
waves became greater and the muck was replaced by sand beaches,


which at this particular point form a strip about 500 meters (one-
third mile) wide and stand about a meter higher than the alluvial
soil farther inland. Tides still run up the rvers, but after crossing
half a kilometer of sandy beach, their effect is mostly limited to
damming up the fresh water of the rivers and the salinity is neces-
sarily greatly reduced. That they still exert some small effect,
nevertheless, is shown by the presence of scattered mangrove trees
along their banks.
Just back from the river bank and at some distance from the
ocean, the conditions are ideal for a normal succession from the
mangroves, since the salinity is greatly reduced by increasing dis-
tance from the open sea and by the continual influx of fresh water
in the rivers. Under such conditions we may expect that the
regeneration of the mangroves by seed will be greatly hampered,
and that they will gradually be replaced by other arborescent
species, adapted to the environment and able to germinate in the
shade. Apparently such is the history of the Pterocarpus forest.
The soil is a pure black muck of unknown depth, standing barely
above water level, always wet and soggy and frequently treacherous
underfoot. The muck becomes shallower toward the edge of the
forest where sand is dug up in planting coconuts. This superposed
muck represents the product of the continued activity of mangroves
or other plants after the sand bars had been formed, and it is alto-
gether probable that its formation is still continuing, due to flooding
by the river during the rainy season and to the continued accumula-
tion of decaying vegetable matter.
Pterocarpus officinalis Jacq. is the dominant tree (Pl. 17) and
constitutes practically all of the arboreal vegetation. Its place in
the upper layer is shared only by a few scattered trees of Roystonea
borinquena Cook and by a very narrow and interrupted strip of
Rhizophora Mangle L. growing immediately along the river. The
trees have been repeatedly cut for fuel at a height of 5-12 decimeters
(2-4 ft.), and coppice growth springing from the stumps has
produced from each tree three to six slender trunks, now reaching
20 meters (60 ft.) in height, but seldom as much as 5 decimeters
(20 inches) in diameter. The old trunks below the coppice growth
are strongly buttressed. In average examples, the buttress roots
are four times as long as high, fairly thin toward their base, but
thickened and sometimes branched distally, with curved upper
edges. They may extend as much as 5 meters (15 ft.) from the


tree and be 1 to 1.5 meters (3-5 ft.) high at the base. Traces of new
development of buttresses may be found as much as 5 meters (15
ft.) above the ground. The size of the buttress system is completely
out of proportion to the size of the trunk; one tree only 3 dm. (1

FIG. 13. Marginal zone of tall Acrostichum aureum, bordering the Ptero-
carpus association at Humacao Playa.
ft.) in diameter at a distance of 4 meters (12 ft.) from the ground
had an enormous development of buttresses measuring 25 meters
(80 ft.) in circumference.
Besides the royal palm, mentioned above, two other arborescent


species are mixed with the Pterocarpus trees, but are always
smaller and relatively uncommon. Tall slender trunks of Drepano-
carpus lunatus (L. f.) G. F. W. Meyer become 1 dm. (4 inches) in
diameter and reach a height of 10 meters (30 ft.). A few young
plants of Clusia rosea Jacq. are anchored in the tops of trees, and
some of them have already sent their roots to the ground and
become independent plants.
The single species of shrub is Malache scabra B. Vogel, well
distributed in small thickets and as isolated shrubs throughout the
forest and reaching a height of 4 meters (12 ft.).
Among the vines, Paullinia pinnata L. is especially abundant
and climbs to the tops of the trees, while its long stems, 5 cm. (2
inches) in diameter, festoon the trunks and extend in loops from
one tree to another. A few plants of Hippocratea volubilis L. also
occur; Aristolochia trilobata L., Banisteria laurifolia L., and Stig-
maphyllon puberum (L. C. Rich) A. Juss. are rare.
The irregular surface of the Pterocarpus buttresses and the
accumulation of decomposed debris among the coppice growth on
the stumps afford excellent sites for the establishment of numerous
epiphytic plants. The largest and most abundant is Anthurium
acaule (Jacq.) Schott, whose dense masses of roots, sometimes 3
din. (1 ft.) in diameter and 6 dm. (2 ft.) long, afford rooting places i
in turn for seedlings of the same species and for Peperomia glabella
(Sw.) A. Dietr. A large fern, Polypodium decumanum Willd., is
also common, and a few plants of Nephrolepis biserrata (Sw.)
Schott grow on the stumps of Pterocarpus. Hylocereus trigonus
(Haw.) Safford is occasional on the higher branches of the trees.
The single herbaceous species is Acrostichum aureum L., growing
as isolated plants and small colonies throughout the forest wherever
the canopy is open enough to admit sufficient light.
At the margin of the association some attempt has been made to
plant coconuts and a strip of the forest has been cleared. Its
place is now taken by an extraordinarily dense zone (Fig. 13) of
Acrostichum aureum L. Its clumps stand close together and its
fronds, reaching a length of giore than 3 meters (10 ft.) are inter-
laced in every direction, forming an efficient natural hedge about the
association. Outside of this zone of ferns, the mucky soil beneath
the coconut trees is carpeted with mats of Bramia Monnieri (L.)
Drake and Nelsonia brunellioides (Lam.) Kuntze.


In the Pterocarpus forest it is obvious that there has been a
progressive decrease in salinity. We may expect this decrease to
continue, as fresh deposits of alluvium raise the land higher above
tidal influence, until the land is eventually similar in environment
to the playa lands of the north coast. We infer accordingly that
the now extinct mesophytic forest of the playa lands will normally
succeed the Pterocarpus association.

We have shown that on shores with active wave action and free
exposure to wind, sand dunes are formed, vegetation is excluded
from the wet fore-shore, and that the initial stage in vegetation is
a xerophytic dune association. Between such conditions and the
quiet shores with mangrove vegetation there is necessarily every
intermediate stage. As wave action becomes less, finer materials
are deposited, plant life is physically possible at lower levels, the
salt content of the soil is increased, and the initial vegetation
becomes more and more halophytic. Such an intermediate stage
is well illustrated at Fajardo Playa, where several offshore islands
reduce wave action far enough to allow mangroves to grow at
the very edge of the water. The waves are still strong enough
to carry up and deposit fine sand, which is raised but a few deci-
meters above the water level and is not blown up into dunes.
Inland there is a gradual change from saline to fresh conditions,
as rainwater leaches out the salt.
The actual beach has a very gentle slope and is almost completely
overgrown by Rhizophora Mangle L., Laguncularia racemosa (L.)
Gaertn. f., and Avicennia nitida Jacq. Just back of the beach the
three mangroves still continue, together with Coccolobis uvifera
(L.) Jacq., Dodonaea viscosa Jacq., and Pictetia aculeata (Vahl)
Urban, forming a dense thicket about 6 meters high, with almost
no herbaceous vegetation. Farther inland, Rhizophora Mangle L.
disappears completely, the other two mangroves become less
abundant, and Conocarpus erecta L. appears in quantity. Coccolo-
bis uvifera (L.) Jacq. and Pictetia aculeata (Vahl) Urban continue,
and these three species together are the dominant members of
a thicket vegetation (Pl. 18) which extends over several acres back
of the beach. Other common species are Randia mitis L., Pithecel-


lobium Unguis-cati (L.) Mart,, Mimosa Ceratonia L., Rhacoma
Crossopetalum L., Ginoria Rohrii (Vahl) Koehne, and Colubrina
Colubrina (Jacq.) Millsp.
The thicket is undoubtedly cut over freely for fuel, but the
vegetation and the environment is still sufficiently similar to the
beach thickets at Mameyes to suggest the opinion that they occupy
the site of a former mesophytic forest, probably equivalent to that
of the white sands already described. We see, then, that the
halarch series, depending for its completion on the maintenance of
salinity, may in a more sandy soil be replaced by members of the
xerarch series and the vegetation carried to a corresponding con-
This great swamp, the largest on the north shore, occupies a
tract nearly 20 kilometers (12 miles) long and from one to three
kilometers (1-2 miles) wide, extending from Arecibo on the west
nearly to Barceloneta on the east. Its natural outlet is into the
Arecibo river at its western end, but artificial ditches intended to
drain the swamp have also been opened into the Manati river at
the east. Much of its surface is actually below sea-level, and none
of it rises more than a few decimeters above tide. The gradient
of the surface is too low to admit of successful drainage and the
opening of ditches and canals has facilitated the entrance of tide
water and thereby carried saline conditions some distance into
its interior portion. The degree of salinity varies notably from
place to place because of the slow diffusion of salt water through
the heavy vegetation. Each rise of the tide brings in salt water,
and each fall of rain tends to wash it out. The clearing of the
original vegetation in some parts of the swamp and the consequent
exposure of the soil in cane cultivation has also tended to increase
the salinity at the surface by the rise of water from below and the
deposition of salts by evaporation, so that the surface is frequently
whitened by saline incrustations.
The soil is a jet-black muck, formed by the accumulation of
silt and the partial decomposition of swamp vegetation. The
water-table stands above or but little below the surface, and the
water increases in salinity from east to west. The natural vegeta-
tion indicates that the soil-water was originally nearly or quite
fresh over the eastern two-thirds of the swamp, while the decided


change in plant life in the western fourth and the presence there of
permanent standing water with direct connection to the sea
through the Arecibo river show saline conditions. The soil is very
rich in organic matter and deficient in mineral substances, particu-
larly in potash.
Viewed from the coastal hills along its northern side, the swamp
presents at its eastern end a vast level expanse densely and com-
pletely covered with narrow-leaved sedges, grasses, and cat-tails,
with a few widely isolated shrubs and small trees. About a fourth
of its length from the western end the vegetation changes abruptly
into a similarly dense expanse of dark green ferns, with scattered
but more numerous trees. Still farther to the west, and extending
to its end at Arecibo, the trees become more numerous and finally
confluent into actual forests, separated still by strips of ferns.
The whole expanse is intersected by numerous artificial canals and
ditches of open water.
These three plant associations occupy almost all of the swamp.
A fourth association of swamp forest originally existed at the
eastern end, but has been completely destroyed and the land used
for cane cultivation. A fifth association has developed in the
canals and drainage ditches, and a sixth type of vegetation, scarcely
to be dignified by the term association, appears on the low ridges of
earth excavated in digging the canals and on the terreplens con-
structed in attempts to cultivate cane.
The Typha-Mariscus association (Pl. 14) occupies almost all of
the eastern three fourths of the swamp. It consists primarily of
a dense mass of narrow-leaved hydrophytes, growing so closely
together that few other plants are able to exist beneath them.
Three species of plants are dominant, but of different order of
abundance. Typha angustifolia L. is the most important, and
covers hundreds of acres with an almost pure stand. Its crowded
stems are 2-3 m. (6-9 ft.) tall and its fallen lower leaves also
contribute to shade the soil and exclude other plants. Second in
quantity is Mariscus jamaicensis (Crantz) Britton. It forms
similar solid masses, its lower leafy stems rising about a meter,
and overtopped by the sparsely leafy flowering stems which attain
a height of 2-3 m. (6-9 ft.). At wider intervals are small thickets
of a bamboo-like grass, Phragmites Phragmites (L.) Karst., growing
in similar crowded masses, its densely leafy stems attaining a
height of 4-5 m. (12-15 ft.) and terminating in a large panicle of


flowers. In general these three species form pure stands, but
scattered individuals of each species occur occasionally in the
colonies of the others.
The secondary species are few in number of individuals and
generally widely scattered. The commonest of them is Pluchea
odorata (L.) Cass., a shrub 1-2 m. (3-6 ft.) tall. This species
alone is tall enough to bring its leaves above the general mass of
foliage, but Mikania congesta DC. and an Ipomoea occasionally
climb over the taller plants and thus secure the necessary light.
In the few small open places between the dominant plants are
numerous plants of Sagittaria lancifolia L., Eleocharis interstincta
(Vahl) R. & S., Hydrocotyle verticillata Thunb., Dichromena colorata
(L.) Hitchc., Bramia Monnieri (L.) Drake, and other species of
the mud-bank association. Occasional plants of Acrostichum
aureum L. at the margin of the larger ditches and scattered in-
dividuals of Conocarpus erecta L. indicate the introduction of saline
conditions by attempts at drainage. A few shrubs of Tabebuia
pallida Miers, seldom more than 2 m. (6 ft.) tall, may represent
relics of the earlier swamp forest.
About three kilometers (2 miles) from the western end of the
swamp the vegetation changes abruptly along a north and south
line across it. To the east there extends the great area of Typha
and Mariscus; to the west they are replaced by similar continuous
masses of the tall fern, Acrostichum aureum L., covering some
hundreds of acres with an almost pure stand. This fern, charac-
teristic of brackish water, grows in large clumps of numerous tall
fronds, reaching here a height of 2 or even 3 meters (6-9 ft.),
and the clumps are so close together that no other plant was
detected which seemed to be a typical secondary species. Scattered
individuals of the thorny legume, Drepanocarpus lunatus (L. f.)
G. F. W. Meyer, with ascending, sparsely branched stems, rise
2-3 m. (6-9 ft.) above the general level, and seem more abundant
in this association than in the mangrove thickets still farther west.
A few plants of Conocarpus erecta L. grow 2 to 3 meters (6-9 ft.)
high among the ferns, and, as previously mentioned, occur also in
limited numbers in the preceding association. A few plants of the
Typha-Mariscus association persist here and there, including small
colonies of the dominant species and patches of Phragmites, but
these are always more abundant on the low ridges along the canals
than elsewhere.


At the western end of the swamp, where the salinity is increased
by the regular entrance of the tides, there is still an extensive
development of the mangrove association. The three dominant
species and the usual secondary species are all represented. Even
in this saline environment a few colonies of Typha and Phragmites
still occur. This swamp forest has been reduced in area by cutting
and probably extended originally much farther to the east over
ground now occupied chiefly by Acrostichum aureum L.
The opening of the numerous ditches and canals (Pl. 19) has
produced areas of standing water so nearly fresh that several
fresh-water hydrophytes ,have appeared in them. The most
abundant is Castalia ampla Salisb., which now covers extensive
portions of the canals with its floating leaves and showy white
flowers. Associated with it are smaller scattered colonies of
Potamogeton fluitans Roth, with smaller but floating leaves, while
submerged plants are represented by Ceratophyllum demersum L.
The term terreplen is used to designate the low flat ridges which
rise 3-6 dm. (1-2 ft.) above the normal level of the swamp, formed
from earth excavated in dredging or thrown up to make roads
used in cane cultivation. Such a condition is entirely anomalous
in the swamp, and comparatively few swamp species are able to
adapt themselves to it. This is particularly true along the larger
canals, where the dredging has been so deep as to bring to the sur-
face the underlying soil produced by decomposition of the lime-
stone. Such terreplens are very sparsely vegetated (Fig. 14),
even after a lapse of 16-18 years. On terreplens of swamp soil the
salvia, Pluchea odorata (L.) Cass., has become common and far
more abundant than in the swamp, due to lack of competition
against the dominant cattails and sedges. Various other species
are of course represented, but usually by widely scattered individ-
uals. The chief vegetation is composed entirely of adventitious
weedy species, among which Borreria verticillata (L.) Meyer is
most abundant. Other species are Ecastophyllum Ecastophyllum
(L.) Britton, Scoparia dulcis L., Solanum torvum Sw., Calophyllum
antillanum Britton, Ficus laevigata Vahl, Corchorus hirtus L.,
Annona glabra L., and Wedelia trilobata (L.) Hitchc.
If the vegetation were still in an entirely natural condition, it
is apparent that the four plant associations would lie in a row, the
mangrove association at the extreme western end, and the fern, the
cattail, and the swamp forest in order toward the east. The


geological and physiographical history of the region also makes it
apparent that the swamp first developed near the Arecibo river at
its western end, and gradually extended farther to the east. This
does not mean, however, that the swamp forest first appeared at the
west end, and gradually pushed to the east following the develop-
ment of the swamp, at the same time giving way to other associa-
tions at the west. On the contrary, we infer that the first swamp
vegetation was the mangrove association, which appeared around
the margin of the young lagoon, as soon as saline conditions were
developed around the mouth of the Arecibo river. Mangroves are
plants of salt but quiet water, have rapid means of dispersal, and

FIG. 14. Caflo Tiburones: the terreplen produced by the excavation of the
canal at the left is occupied sparsely by weedy species.

appear promptly after the establishment of conditions suitable to
As the swamp extended in size and the mangrove thickets
became larger, they served to break the tides and to prevent its
complete mixture with the fresh water flowing in from the east.
In this zone of quiet water, ranging from brackish to nearly fresh,
the other three associations established themselves as narrow zones.
By further extension of the swamp lands, due perhaps to progres-
sive submergence of the land, all of the plant associations increased
greatly in size, but this increase affected the Typha-Mariscus associ-
ation most, and it has come to dominate most of the swamp at the
present time. The early stages in such a development may now be


seen at Palmas Altas, where a small area of swamp is now establish-
ing a mangrove thicket. A variant of this history occurs at San
Juan harbor, where the deeper water, the better exposure to the
tides, and the steeper gradient of the surrounding land, has led to a
great development of mangroves and a corresponding restriction of
the Typha-Mariscus and Acrostichum associations.
The present comparatively sharp transition between the man-
groves and the Acrostichum, and between the Acrostichum and the
Typha-Mariscus associations is not caused by a similarly sharp
transition in the salinity of the water. That decreases gradually
from west to east and would produce an equally gradual change in
the vegetation if it were the sole cause of the vegetational dif-
ferences. Besides, the tolerance of the typical plants is greater
than exists in the areas occupied chiefly by them, so that scattered
colonies of Typha are found among the mangroves, and scattered
plants of Acrostichum among the Typha and even in immediate
proximity to the canals with typical fresh-water plants such as
Castalta ampla Salisb. The abrupt change is due to the activity
of the vegetation itself, both Acrostichum and Typha occupying
the ground so completely that mixture of the two associations is
difficult and has led only to the establishment of isolated colonies.
It is probable that this close control of the environment is due
chiefly to shading of the surface so that seeds or spores can not
germinate. This solution is almost certainly true for the man-
groves, which shade the soil so densely that secondary species are
virtually absent.
The evidence for future changes in the vegetation is very slight.
It is obvious that there are more colonies of Acrostichum in the
Typha-Mariscus association than the reverse, and this may indicate
an advance of the Acrostichum association farther to the east.
Since the control of light conditions by the mangroves is so com-
plete, it is conceivable that the isolated plants of the various
mangroves among the Acrostichum may develop into larger
thickets and accordingly extend the mangrove association eastward
to the limit of its tolerance of fresh water. It is also possible that
much or all of the area of the present Acrostichum association was
originally occupied by mangroves, which have been destroyed by
cutting for fuel.
The whole area of the great swamp is nearly surrounded by
fields of cane, which have been extended into the swamp as far


as the limit of tolerance of the cane will permit it to grow. Attempts
have been made to extend cane cultivation still farther by throwing
up the soil into small elevations separated by narrow canals, but
here the rise of salt by capillarity and its accumulation in the
surface soil by evaporation of the water have often led to the aban-
donment of the trial. On such deserted plantations (Pl. 20), the
canals are very quickly occupied by Typha and Mariscus, the edge
of the cane beds by plants of the mud-bank association, particularly
Bramia Monnieri (L.) Drake, and the surface of the beds by various
weedy species, among which Wedelia trilobata (L.) Hitchc. is
always prominent.
The pioneer association of the halarch series is a mangrove
swamp, developed in saline conditions at the edge of the shore. As
soil is accumulated, the landward margin of the mangrove forest
becomes less saline, and is finally succeeded by a forest of Ptero-
carpus. It is probably that this in turn is succeeded, under strictly
natural conditions, by the climax forest of the region. Clearing
the mangrove association leads to the development of an association
of Acrostichum aureum L., which tends to be replaced by the man-
groves. In situations intermediate in environment between the
mangrove and the Coccolobis associations, mixtures of species from
both associations occur. In some of the larger lagoons, where areas
more remote from the ocean are characterized by fresh or slightly
brackish water, the vegetation is replaced by associations of the
hydrarch series.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs