Full text of "A naturalist in the Bahamas, John I. Northrop, October 12, 1861-June 25, 1891; a memorial volume"

See other formats

ICNJ

ICO
ICO

loo

^co

,^^

y '

LIFE AND WRITINGS OF JOHN I. NORTHROP

A NATURALIST IN THE
BAHAMAS

JOHN I. NORTHROP

October 12, 1861 — June 25, 1891

H flDemorial IDoIume

EDITED WITH A BIOGRAPHICAL IxNTRODUCTION

HENRY FAIRFIELD OSBORN

Mi).

THE COLUMBIA UNIVERSITY PRESS
1910

Published June, 1910.

J. S. Gushing Co. — Berwick & Smith Co.
Norwood, Mass., U.S.A.

IN ADDITION TO THE PAPERS BY DOCTOR NORTHROP, THE VOLUME
CONTAINS AN ACCOUNT OF THE FLORA AND NARRATIVE OF THE
BAHAMA TRIP BY MRS. NORTHROP ; A REPORT ON THE CRUSTA-
CEANS OF THE COLLECTION, CONTRIBUTED BY PROFESSOR WILLIAM
H. RANKIN OF PRINCETON UNIVERSITY ; A REPORT ON THE ACTINIA,
CONTRIBUTED BY PROFESSOR J. PLAYFAIR McMURRICH OF THE UNI-
VERSITY OF TORONTO, FORMERLY OF CLARK UNIVERSITY ; A LIST OF
THE SHELLS, AS DETERMINED BY PROFESSOR WILLIAM H. DALL OF
THE UNITED STATES NATIONAL MUSEUM ; AND A DESCRIPTION OF
ICTERUS NORTHROPI BY PROFESSOR J. A. ALLEN OF THE AMERI-
CAN MUSEUM OF NATURAL HISTORY. OF THE BOTANICAL COLLEC-
TIONS, THE MARINE ALG^ WERE DETERMINED BY MR. FRANK S.
COLLINS OF MALDEN, MASSACHUSETTS, WHILE THE ACCOUNT OF
THE PALMS WAS CONTRIBUTED BY DR. O. F. COOK OF THE DE-
PARTMENT OF AGRICULTURE AND THE UNITED STATES NATIONAL
MUSEUM.

OTHER ASSISTANCE AND COOPERATION IS DULY NOTED IN THE
SEPARATE PAPERS THROUGHOUT THE BOOK.

TABLE OF CONTENTS

PAGE

Introduction xi

Henry Fairfield Osborn.

Bahaman Trip, General Notes i

Alice R. Northrop.

Notes on the Geology of the Bahamas .27

John I. Northrop.

Bahaman Birds 48

John I. Northrop.

The Birds of Andros Island, Bahamas 50

John I. Northrop.

Description of a New Species of Icterus from Andros Island,

Bahamas 66

J. A. Allen.

The Northrop Collection of Crustacea from the Bahamas . 69
W. M. Rankin.

List of Shells collected by Dr. John I. Northrop in the Bahamas 99
Identified by Professor William Healey Dall.

Notes on Some Actinians from the Bahama Islands collected
BY THE Late Dr. J. I. Northrop. With Appendix of

Date of 1908 103

J. Playfair McMurrich.

Flora of New Providence and Andros (Bahama Islands) . .119
Alice R. Northrop.

Cultivation of Sisal in the Bahamas 212

John I. Northrop.

The Eruption of Krakatoa in 1883 225

John I. Northrop.

vii

viii TABLE OF CONTENTS

PAGE

Plant Notes from Temiscouata County, Canada .... 240
John I. Northrop.

Plant Notes from Tadousac and Temiscouata County, Canada . 250
John I. and Alice R. Northrop.

Notes on the Distributioj^ of the Plants of Mount Washing-
ton, N.H 255

John I. and Alice R. Northrop.

A Study of the Histology of the Stem of the Wax Plant,

HovA Carnosa (L.) R. Br 259

^ John I. Northrop.

LIST OF ILLUSTRATIONS

Portrait of Dr. John I. Northrop . . . . . . Frontispiece

PAGE

Ryswick ........... Facing 3

The Bay at Nicol's Town " 6

House at Nicol's Town ........ " 8

At Conch Sound . " 14

Fresh Creek " 19

Cabbage Creek " 20

Coming to Anchor off the Dames Place ..... " 22

Goat Cay, Andros ......... " 32

Ocean Hole, Nicol's Town " 34

Spruce Cay, New Providence, showing Erosion .... '' 40

Icterus northropi Alien . . . . . . . . '' 66

Crustaceans . . . . . . . . . . . -97

Crustaceans 98

Actinians from the Bahama Islands ...... Facing 116

Map of New Providence and Andros " 118

Main Street in Nicol's Town ....... '' 122

In the High Coppet near Deep Creek " 124

Aletris bracteata ......... " 144

Hymenocallis arenicola ........ " 146

Vanilla articulata " 148

Phoradendron northropice . . . . . . . " 150

Pithecolobium bahamense . . . . . . . . " 155

Cassia caribcea .......... " 156

Linutn bahainense . . . . . . . . . " 159

Erythroxylon reticulatum ........ " 160

Crossopetalum coriaceum ... . . . . . . '' 165

Reynosia northropiana ........ " 166

Helicteres spiralis ......... " 167

Xylosma ilicifolia . . . . . . . . . *' 169

Terminalia spinosa ......... '< 172

Metastelma barbata " 176

LIST OF ILLUSTRATIONS

Heliotr opium nanum ......... Fac\

Tecoma bahamensis .........

CatesbcBa fasciculata' .........

Anguria keithii ..........

Myrstiphyllum ligustrifoliimt .......

Eupatorium bahamense ........

PAGB

182
184
186
187
188

Swash, West Side of Andros 218

Clearing the Pine-yard for Sisal near Nassau, N. P. ..... 219

The House of a Sisal Planter, Andros . . . . . . .221

Map of the Islands of the Krakatoa Group before Eruption of August, 1883 227
Section through Length of the Island of Krakatoa ..... 229

Krakatoa after Eruption of August, 1883 . . . . . . .231

Outline of Crater of Krakatoa . . . . . . ... 232

Hoya carnosa (L.) R. Br . 274

Hoya carnosa . . . . . ... , . . . 277

INTRODUCTION

John I. Northrop was born in New York City October 12, 1861.
He was named after his father, John Isaiah Northrop, born near
Rochester, N.Y., a pharmacist. His mother, Mary R. Havemeyer,
was a sister of Frederic C. Havemeyer, a graduate of Columbia Col-
lege, after whom Havemeyer Hall is named. His father died when he
was two years old.

Dr. Northrop studied for some years at a private school in New
Windsor, N.Y., then at the Columbia Grammar School, in which he
prepared for the Columbia School of Mines. He graduated with the
class of 1884, with the degree of Engineer of Mines. While in college
he was very fond of outdoor life, was a member of his foot-ball team,
an enthusiastic fisherman, and interested in all outdoor sports. One
of his former classmates narrates how he once saved a companion's
life at the risk of his own. They were ascending a shaft at one of the
Lake Superior mines on a man engine when his companion, some dis-
tance above, missed a step and was falling down the shaft. Young
Northrop seized him as he fell and succeeded in holding him with one
arm while he supported himself with the other, an act which called
for a strong arm, quick action, and steady nerve, and showed that
complete forgetfulness of self which was ever one of his most prom-
inent characteristics.

Immediately after graduation he accepted a position in Have-
meyer & Elder's Sugar Refinery, but shortly afterward he journeyed
to Deadwood, So. Dak., where he joined a former classmate, and
together they opened an office as "mining engineers, chemists, and
assay ers." He returned to New York in the spring of 1886 because
of the serious illness of his mother, whose death occurred shortly
afterward.

It was evident that his tastes did not lie in the direction of engineer-
ing, and while the circumstances of his coming back to New York
were sad, they proved to be the turning point in his career. He

xii INTRODUCTION

immediately came under the personal and most inspiring influence of
Professor John Strong Newberry, the veteran teacher with whom he
had studied geology and palaeontology in Columbia College while pre-
paring for his degree. He greatly admired Dr. Newberry, who perceived
his talents and showed great confidence in his ability, and who also
was warmly attached to him, treating him like "one of his own boys."
On January 9, 1888, he was appointed Honorary Fellow in Geology in
Columbia College. Seizing every opportunity which the museum and
his close acquaintance with Professor Newberry afforded, young
Northrop rapidly made himself familiar with the principles of botany
and zoology. He exhibited that breadth of interest which was such a
marked characteristic of his master; but made a specialty of zoology,
and delivered a series of lectures in this subject while holding his
fellowship. He became a candidate for a higher degree, and in 1888
was awarded the degree of Doctor of Philosophy on the presentation
of his two dissertations: "Histology of Hoya carnosa'^ and "Fossil
Leaves from Bridgeton, N.J." The work on the Fossil Leaves from
Bridgeton was carried on under Professor Lester F. Ward of the U. S.
Geological Survey at Washington, and was ultimately to be published
at Washington by the government. Previous to this he had published
"Plant Notes from Temiscouata County, Canada," in the Bulletin
of the Torrey Botanical Club, November, 1887.

After receiving his degree, in the summer of 1888 he accompanied
Dr. Newberry to Colorado and assisted him in some geological work.

On June 28, 1889, he married Alice Belle Rich, at the time Tutor
in Botany at the Normal College, New York City, a companion who
shared all his scientific tasks and the value of whose assistance in all
his work he never failed to appreciate and acknowledge.

Together they spent much of the summer of 1889 at Eastport,
Me., and Grand Manan, N.B., dredging, shore collecting, and
studying marine invertebrates. Dr. Northrop was an enthusiastic
and indomitable collector and most careful observer, sparing neither
time nor trouble to complete or render correct whatever he under-
took. He never could stifle his sense of humanity through his enthu-
siasm as a naturalist. The following, as an illustration of his love of
animals, is recited by Dr. Arthur Hollick : While at Grand Manan,
during the summer of 1889, he visited one of the little islands which
the petrels had long used as a breeding-place. He took several of the

INTRODUCTION xiii

young birds out of their holes in the ground, examined them carefully,
and, to the surprise of his companion, returned them, not taking even
one for a specimen, although the birds were comparatively rare, and he
was not likely to have such an opportunity again. He explained that
he could not make up his mind to take one of these birds because he
realized that each mother bird had but one little one all summer. He
used to laugh afterward at what he called his "foolishness," but this
consideration for animals was one of his most charming characteris-
tics. .

In the early part of the year 1890 a journey to the Bahamas was
planned. The islands of New Providence and Andros, especially, were
visited and studied carefully. The amount and variety of material
collected during this trip, which extended from January 2 to July 10,
was remarkable; geology, botany, and zoology were all represented
both in the collections and in the notes. From observations made on
this trip were published his papers ''Notes on the Geology of the
Bahamas," "The Cultivation of Sisal in the Bahamas," "Birds of
Andros Island, Bahamas." Among his unfinished works almost
ready for publication were: "A List of Starfishes and Ophiurans col-
lected in the Bahamas," and "A List of Sea-anemones collected in
the Bahamas."

One of the birds collected on this trip proved to be new to science,
^nd was described and figured by Professor J. A. Allen in the Auk.
January 8, 1891, under the name Icterus northropi. The sea-anemones
were subsequently handed over to Professor J. Playfair McMurrich
for description.

The report on the plants brought together on this trip was pub-
lished subsequently by Mrs. Northrop and forms part of this volume.

The autumn and winter of 1890 Dr. Northrop devoted to the study
and preparation of these collections. In the meantime his abihty was
widely recognized. He held membership in the Torrey Botanical
Club, in the American Institute of Mining Engineers, in the Linnaean
Society of New York, in the American Ornithologists Union, in the
American Folklore Society, in the American Association for the Ad-
vancement of Science, and in the New York Academy of Sciences.
He also was enrolled as a member of the International Geological
Congress during the 1888 meeting.

During the winter of 1890 President Low and the trustees of

xiv INTRODUCTION

Columbia decided to greatly extend the zoological courses of the
college, and Professor Henry Fairfield Osborn was invited from Prince-
ton to plan a Department of Biology for the institution. The result
was the creation of a strong department, including a number of pro-
fessors and instructors. On the 5th of May Dr. Northrop was ap-
pointed Tutor in Zoology in this new department, especially in charge
of invertebrates. In order to perfect himself for the duties which he
expected to assume he obtained leave of absence for a year and in-
tended to start during the early part of September, 1890, for Germany
to study under Professor Her twig in Berlin and Professor Haeckel in
Jena.

At the very opening of this bright outlook for the future and attain-
ment of this splendid preparation, in which Dr. Northrop had shown
such rare ability and resourcefulness, came the tragedy of his sudden
death, connected with the performance of his duties as curator of the
zoological collections in the School of Mines Museum.

In the afternoon of June 25 Dr. Northrop had occasion to obtain
some alcohol for use in the preservation of zoological specimens. As a
measure of safety the main supply was stored in a fire-proof vault
under the School of Mines building. Two janitors accompanied him,
and together they proceeded to draw off some of the alcohol from a
large cask into a smaller vessel. The vault was dark, and a match was
struck in order to see how the vessel was filling. Just what happened
we shall never know ; but the alcohol ignited, there was a momentary
glare of flame around the faucet, a flash, an explosion, and a sea of fire
in the vault. All three of the men were thrown down and badly burned,
but evidently Dr. Northrop' s clothing had become saturated with the
alcohol, and his injuries were so severe that he died during the follow-
ing night. On July 5, 1891, just ten days after Dr. Northrop's death,
his son, John Howard Northrop, was born.

It is difficult to measure the loss to Columbia College, so soon
(1896) to become Columbia University and to assume its important
part in the scientific work of America. His memory will be per-
petuated in Columbia by the Zoological Library which he was in-
strumental in getting together, through the generosity of his uncle,
Mr. Charles H. Senff, and in the zoological material which he col-
lected and arranged for the Zoological Museum.

The finished and unfinished work which Dr. Northrop left behind

INTRODUCTION xv

him gave evidence of acute powers of observation, of painstaking
study, and of strict regard for truth in the recording of facts, qualities
which are the prime essentials of a successful man of science. Per-
haps the most notable characteristic of his mind was his keen ap-
preciation of underlying causes in the investigation of facts. His
excessive modesty gave him an appearance of reticence and reserve
which was not always understood by those who did not know him
intimately. To his personal friends this reserve was merely one of
the evidences of his high motives and scientific idealism. His ideals
were further manifest in the indifference with which he regarded the
discovery and description of new species as compared with the study
of their underlying structure and life history. He was inspired by a
passion for scientific work and also a passion for truth, which, with
the greatly enlarged opportunities opening toward him, would surely
have enabled him to leave a permanent mark on American science.

Dr. Northrop had in mind the ultimate publication of a volume on
the Bahamas which would embody the results of his work there and
of other contemplated collecting trips to the islands, and this memorial
volume was planned to represent the carrying out of that project of
his so far as possible. The contemplated work would have been far
more complete and more worthy of the author ; but we trust that this
volume, which includes all of his writings, largely brought together
through the loving care of Mrs. Northrop, will form a permanent
memorial of John Isaiah Northrop, so that the memory of his labors
and activities, brief as they were permitted to be, may live, and the
influence of his example be handed down to future generations of the
students of Columbia and of other universities.

Henry Fairfield Osborn.

BAHAMAN TRIP

General Notes
By Alice R. Northrop

During the summer of 1889 Mr. Northrop collected and studied
the marine invertebrates of northern waters at Eastport, Me.,
and Grand Manan, N.B. He was anxious to follow this up with
the study of southern forms preparatory to taking a place upon
the teaching staff of the newly organized Zoological Department of
Columbia University. Owing to the wealth of their marine life, a trip
to the Bahama Islands was decided upon, and when we found that
their flora was only imperfectly known, it was planned to make a col-
lection of the plants also. A leave of absence was secured and over six
months were spent on the islands, two on New Providence and the
remainder of the time on Andros. The scientific results of the trip,
as far as the material has been worked up, are given in the following
papers. Below is a brief account of our journeys and such general
information as has found no place in the special papers. The follow-
ing narrative is compiled almost wholly from my husband's note-
books and well shows his keen powers of observation and the pains-
taking accuracy of his notes.

We left New York on Thursday, January 2, 1890, and at dawn on
the following Monday the steamer dropped anchor off the city of
Nassau. We found it a picturesque place, with white roads bordered
on either side by low-roofed houses, often embowered in vines and set
in pleasant gardens. " The main thoroughfare and principal business
street is Bay Street, running parallel to the water and extending east
and west for several miles. A short walk to the west brings us out of
the town. The roads and gardens are bordered with walls of coral
rock, plastered all over or often on top only. In the parks and gar-
dens and planted along the streets are cocoanut-palms, almond trees
(Terminalia), with their dark green glossy leaves, Spanish cedars (Cas-
uarina), the sand-box tree {Hura), and the silk-cotton tree or ceiba,

B I

2 BAHAMAN TRIP

with its spreading horizontal branches and buttressed trunk. The
silk-cotton trees that have leaves show but few flowers, and the flower-
ing branches are leafless. The trees that have flowers are one mass
of buds. Although the silk-cotton tree is a native of more southerly-
islands, it flourishes on New Providence and is quite common about
Nassau, as is also the beautiful ''flamboyant tree" {Poinciana regia)^
from Madagascar. The latter has a dome of finely dissected, fernlike
foliage, amid which, at the time of our arrival, hung the curious brown
pods, two feet or more in length. On Andros, in June, the Poincianas
were a mass of scarlet flower clusters, and as the leaves were not yet
developed, the dome of flaming scarlet could be distinguished several
miles from shore.

Almost all the inhabitants of New Providence live in Nassau and
its environs, there being only a few small negro settlements on other
parts of the island. In Nassau itself, said to have a population of
about fourteen thousand, probably five-sixths of the inhabitants are
negroes. These live principally in the suburbs known as Grant's
Town, Bainville, and Fox Hill. Here their huts are close together
and are generally surrounded by little gardens of cocoanuts, orange
trees, and bananas. These Bahaman gardens look strange to North-
ern eyes, for the vegetation apparently springs from bare coral rock.
The surface, however, is much weathered and very irregular, and the
scanty soil and leaf-mould that collects in the holes is sufficient to
support the luxuriant vegetation."

The chief boast of the people of Nassau is their climate, and after
our six months' sojourn we concurred in their opinion that most of
the time it was as nearly perfect as a climate could be. While we were
on New Providence, the average temperature was about 75 degrees,
from which point it seldom varied. From records kept by my husband
on Andros, the greatest heat was recorded at Lisbon Creek on June
23, when it was 86 degrees indoors and 125 degrees in the sun. We
had one "norther" while on New Providence, when the wind blew a
gale for several days, raising a heavy surf on Hog Island and Salt Cay.
The waves broke over the latter in places where it was thirty feet in
height. This was the only storm of any account experienced during
the entire trip. During the first four months there was very little
rain. Later in the year heavy showers were frequent, although there
was rarely a day on which the sun did not shine for a time, at least.

Ryswick.

BAHAMAN TRIP

The greater part of our sojourn on New Providence was spent at
*'Ryswick," a country place which we rented of one of the mer-
chants in town. It was about three miles east of Nassau and near
the best grounds for marine collecting. The house was situated on
the rocky coastal ridge and commanded a fine outlook. The gray
walls of old Fort Montague were on one side and Dix Point on the
other, while we looked across the bay to the low, wooded cays known
as Hog Island and Quarantine. Beyond these could be seen the long
narrow ridge of Salt Cay. A man and sail-boat were engaged, and
day after day we sailed to the outlying cays and reefs and collected
along their shores. The marine life was so wonderfully varied and
abundant that for several weeks we daily brought back starfishes,
holothurians, crustaceans, corals, or sponges we had not found before.
Dix Point and the sand flats in front of the house also proved profit-
able collecting grounds at low tide. Occasionally we would try dredg-
ing, but the coral heads and blocks made it difficult work, and we
did not find it as productive as shore collecting. What is termed the
"sea-garden" is near Ryswick. Here one looks down into a forest of
alcyonoid corals, pink, brownish, and yellow, while scattered over the
bottom were large sponges, some round and inky black, others clus-
tered and purple-tinted, also huge brain corals interspersed with
many of the delicate branching madrepores. Gaily colored fishes
darted about, the most striking being bright blue, while others
showed yellow bodies and blue tails or were silver spotted with crim-
son.. The water was so wonderfully clear that it was difficult to be
convinced that these marine treasures were not within easy reach.
We also visited the living coral reef off Rose Island, some miles to
the northeast, but the reef we saw later along the east coast of Andros
was much larger and finer.

The principal work at New Providence was the collection of
marine invertebrates, but between times, and when it was too rough to
collect, we explored the island in many directions, visiting the south
side, the southwest beach, the extreme eastern end as well as Lake
Cunningham and the caves to the west of Nassau. Two or three
hundred species of plants were collected on these trips as well as some
birds and insects.

January 1 1. Drove across to south side of island. The north side,
just back of the town, is a rocky ridge; then comes a comparatively

4 BAHAMAN TRIP

level space containing cocoanut plantations and a small patch of sisal
hemp. In the uncleared lowlands Pinus Bahamensis covers the
ground. The higher parts are covered with angiosperms. Near the
water the land is very low, the soil soft and calcareous. The only
vegetation seems to be numerous, small, scattered mangroves and a
few buttonwoods (Conocarpus).

February 3. To Lake Cunningham. Followed the road through
part of Grant's Town, then a winding roacj leading through a pine
barren for about five miles in a general southwesterly direction.
Crossed a number of ridges running north and south. Noticed some
large banana holes. The pines have mostly been cut down, and the
ground is covered with a second growth. At the eastern end of the
lake is a mangrove swamp. The lake is perhaps a mile and a half
long and about a half of a mile wide. The water is slightly brackish.
A great number of shells on the bottom, near the shore ; collected some
(determined as Area Jamaicensis by Dr. Dall). The drive home
showed the lake to be quite close to the sea. The country to the west
of the city is quite fiat.

February 10. In evening walked to Fort Montague. Water very
calm. Saw a worm about an inch and a half long swimming through
the water and leaving a bright phosphorescent track. Came back to
house to get glasses and net. Walked out in water. The bottom was
phosphorescent, every step causing it to scintillate for a circle three or
four feet in diameter. Agitation of the surface produced no light.
Here and there at the bottom were small glowing spots which appeared
to be bivalved crustaceans.

February 20. Fishing at sea-garden. The noise made by market
fish and grunts is produced in the following manner: The upper
pharyngeals are drawn together and backward; the lower is pushed
upward and forward, thus grating on the other two. The noise is
slight, resembling that made by the finger-nail in scratching across
the grain of a board.

February 21. Shark fishing to the north of Salt Cay. Caught one
known as a mackerel shark. Color dark steel-gray above and on sides,
as far as line between pectoral and ventral fins, below this a pure
white with only few blotches of gray, to anal fin, behind which the gray
predominates. The nictitating membrane is white and slides over
the eye from below. The teeth are serrated, with a deep notch in the

BAHAMAN TRIP 5

outer side so that the point projects outward. Nostrils are double,
the one nearest the median line being covered by a flap of skin. The
head viewed from above is very obtuse, almost truncate, and the eye
is just visible ; gill openings fine. There are two dorsals ; one, slightly
in advance of half the distance between the pectorals and ventrals,
is about as high as long, with a point projecting behind half as long as
base of fin. Second dorsal is almost opposite the anal fin, but slightly
in advance of it. There is a pit at base of caudal fin, both above and
below. The upper lobe of caudal fin twice as long as lower lobe and
has notch in lower edge near the end. The ventrals are quite close to
the anal, so that the subcylindrical claspers touch the anal. The
pectorals are large and are continued behind near the base. Length
of shark from fork of tail to end of nose, 8 feet 4 inches ; from tip of
tail, 10 feet 4 inches. Parasites taken from shark : (i) A Remora-like
fish; (2) a crustacean found in gill- slit; (3) leechhke parasite from
roof of mouth one and three-quarters inches in length when fully ex-
tended.

March 5. Dug for Balanoglossus in the sand near Dix Point.
Their holes are marked by a pile of castings. The holes extend down-
ward, usually with a curve, to the rock. Followed one of the holes
down straight about ten inches, then along in a spiral for about eigh-
teen inches to the rock, where I found a Balanoglossus. Followed two
more holes, but lost them before reaching the animal. When found, the
whole animal is surrounded by a transparent gelatinous coating. Laid
one on paper; when extended, it is twelve inches in length. Put one
in fresh water, caused strong contraction ; added some alcohol to salt
water, caused contraction and a discharge of mucus, like the white of
an egg. Shortly after added considerably more alcohol.

March 6. Balanoglossus in very bad condition this morning.
They are coated with mucus, and so soft they will hardly hold together.

[July 5. (After our return from Andros.) Went out to Ryswick
and dug more Balanoglossus. Some were over two feet long. When
an animal was laid on paper, the slime was strongly luminous with a
greenish light that was visible within five feet of a strong kerosene
lamp. The light was strongest between the folds of skin near the cen-
ter of the body.]

After two months collecting on New Providence, we found little
new material; the supply of alcohol was nearly exhausted, and we

6 BAHAMAN TRIP

decided to select some other island for our botanical work. Andros
was chosen as being the largest and least known of the group. Al-
though the nearest part of this island is only twenty-five miles from
New Providence, we could get but very meager information concern-
ing it. The Nassauans know scarcely anything of what they rather
contemptuously designate the ''out islands." Every one tried to
dissuade us from going to Andros, assuring us that "there were no
white people, no roads, and nothing to eat." We were finally fortunate
in meeting Mr. Alexander Keith, of Edinburgh, who had just bought
two thousand acres on Andros and started a sisal plantation. He
kindly told us of a little house we might hire at Nicol's Town, the most
northerly settlement. We added a camping outfit to our baggage,
hired a sloop, and then, with all our goods and chattels packed, were
fain to wait ten days on the pleasure of the wind. We had just had a
heavy ''norther," and our captain could not be persuaded to start as
long as the winds were high, the east coast of Andros being fringed
with a dangerous reef and without a single harbor. Finally on
March 14 we crossed the Tongue of Ocean, making the thirty-five
or forty miles in about eight hours. We passed through a narrow
opening in the reef and disembarked at Nicol's Town on a beautiful,
curving, white beach with thatched huts showing here and there
through a fringe of waving palms. The justice of the peace, a stal-
wart negro, welcomed us to Andros and led the line of march to the
"mission house," where we were to stay. A large contingent of the
villagers followed, bearing our boxes and various pieces of baggage
on their heads. The house was only eighteen and a half by ten feet,
but it was divided into two rooms and boasted two doors and six win-
dows. It was what is called a "tabby house," one made of coral
blocks, plastered inside and out. Daylight could be seen here and
there through the shingled roof. The house was prettily situated on a
rise of ground, a little apart from the rest of the settlement. Orange
trees were in bloom just outside the windows, and gum-elemi and
cassada trees shaded the dooryard. The gum-elemi proved a great
attraction to the birds, and there would often be seven or eight on it
at once, representing three or four species. The most frequent visi-
tors were the mocking birds, cat birds, the Tom- James bird {Spin-
dalis zena), and the grassquit (Euetheia hicolor). When we passed
from the coppet to the pine-yard, these were replaced by the blue-

The Bay at Nicol's Town.

BAHAMAN TRIP 7

gray gnatcatcher, the pine warbler, and the beautiful little humming-
bird known as the wood-star.

We made Nicol's Town our base of operations for about seven
weeks, collecting in the immediate neighborhood and also making a
number of more extended trips. As comparatively little seems to be
known of the interior and west side of Andros, the following account
of our journeyings is given in considerable detail. "Nicol's Town. it-
self was a straggling village, having, according to Mr. Davis, Justice
of the Peace, a population of 'three hundred and odd.' There were
three churches, Methodist, Baptist, and Episcopalian, and still another
was in process of construction. There is a schoolhouse and a small
jail on the hill; the latter is said to be generally empty. The houses
here are mostly square and are either built of coral rock like ours or are
of timber with smaller sticks interlaced or ' wattled ' on which the plas-
ter is laid. They have sloping roofs thatched with palms. There is
but one well in the village, and the water in that rises and falls with the
tides. It is very poor, hardly fit to drink; even when boiled it has a
brackish taste. We mostly use jelly-cocoanuts and oranges when we
can get them."

March 26. The weather is cold, thermometer 62 at 6.40 P.M.
A hard northeast wind is blowing. In the morning walked along the
beach north of the village, passing through a grove of cocoanut-palms
with here and there a few houses. Some of the trees had been blown
down, but continued to grow so that while about twenty feet of the
trunk lay on the ground, six or eight feet was growing upwards at
right angles to the rest. As the roots were almost entirely out of the
ground and resting on one edge, and as I did not see any secondary
roots at the bend, it would seem as if the cocoanut required very little
nourishment from the soil. The beach along which we walked for
about three miles was fringed with the sea-grape (Coccoloba uvifera),
back of this for most of the distance were three rows of cocoanut-
palms. The few houses stood back of these, and behind the houses
was "the bush."

March I y. After breakfast shot some birds. We then walked a
short distance through the village, turned south into a path leading
past a few scattered houses, a new sisal field and in about three-
quarters of a mile reached the pine-yard. The path was just wide
enough for single file, and on each side the trees and shrubs made an

8 BAHAMAN TRIP

impenetrable bush. The path was exceedingly rough and at times
passed close to the edge of deep banana holes. In the pines saw some
rain crows {Crotophaga ani). Insects seemed very scarce; caught two
butterflies. The largest pine seen was about twenty-four inches in
diameter and about fifty feet high. The bark is gray, quite rough and
detaches easily in flakes, leaving the exposed part dull, reddish
brown. The branches are near the top of the tree, and the leaves
towards the ends of the branches. The cones are very numerous
and are sessile below the leaves.

March i8. Started with Mr. Davis (the justice of the peace)
to visit some caves near Morgan's Bluff. On the way we called on a
Mr. James, who took us to see his cave, a long, low chamber on the
side of a hill. Some of the openings were about thirty feet in height,
and from the overhanging rock, which was about three feet thick, hung
down the long aerial roots of a Ficus, making great bars across the
mouth of the cave. Near by was what is called the " sea-hole or ocean
hole," perhaps a hundred feet in diameter and from twenty to forty
feet in depth. (See account of the geology.) After visiting this we
continued our walk to Morgan's Bluff at the northern end of the island.
The rock here was quite precipitous, much water-worn and about
fifteen feet in height. The cave which was near here had a small
entrance about the size of an ordinary door. A second opening led
into a larger room in which we could stand erect. Found a human
humerus on the ground. Mr. Colby, the owner of the cave, said that
all the earth had been taken out about thirteen years before, sold for a
dollar a ton and taken to Port Royal, S.C. He also said that he had
collected the bones and buried them again. On the roof of the cave a
number of small stalactites showed that the water dissolved some of
the lime as it trickled through. On the return trip Mr. Davis showed
us what are here called ''cinnamon trees" (Canella Winterana), also
the mahogany and madeira trees. The so-called "road," in reality a
narrow path, was very rough, and we were tired when we reached our
cabin, although we had only walked about eight miles.

March 20. Found some large soldier-crabs under a stone, and this
evening when I went out to open a cocoanut, I saw eight or nine of
them around the place. I watched them feeding on the cocoanut
jelly. They seem largely nocturnal in their habits. Caught and pre-
served two males.

Our House at Nicol's Town.

BAHAMAN TRIP

March 21. Boy brought some lizards. One had its tail broken
off. I placed it on the ground, where it opened its mouth and kept
facing my hand, at the same time distending its gular sac in evident
anger. This shows that the extension of the gular sac is sometimes a
mark of anger, still it is possible that it might have been pain.

March 29. In afternoon walked to Cocoanut Point, about a mile
and a half south. Very warm. Road mostly over the rough rock on
the shore, the rest on sand under the cocoanuts. A little distance
beyond the point were met by Mr. Keith, who sailed us to his sisal
plantation on Conch Sound about five miles away. We had a fair
wind and soon entered the sound, and in a short time were sailing
through a winding creek, bounded on each side by mangroves. The
house stood in a small clearing surrounded by pine trees and a coppet
of smaller hardwood trees. A short walk through the pines took us to
the sisal field. Mr. Keith has cleared about three hundred acres of his
land and has it partly planted. Our path was an old road, where we
could plainly see the marks of wagon wheels, said to have been made
about fifty years before, during the slave time. Near the house is a
large chimney and the remains of a generous fireplace. Many swal-
lows were flitting over the clearing.

March 30. Walked along a path through the woods near the shore
and found a number of plants we had not seen at Nicol's Town. There
were many bromeliaceous plants and a number of orchids, some in
fruit and a few in flower. On the return to Nicol's Town saw in the
shallow water of the creek what was known as the "boiling hole."
(See notes on geology.) There were a number of large fish in the hole.
These are sometimes obtained by pounding up the bark of one of the
native trees {Ichthyomethia piscipula), placing it in a sack, and sinking
it in the hole. This is said to stupefy the fish, which then float to the
surface.

Monday, April 14. Left Nicol's Town in a small sloop, the Her-
ald, for a short trip to the west coast. The Herald was about one ton
burden and drew from one to two feet of water. It was manned by a
negro captain and cook. Mr. Keith accompanied us. Passed the
end of the reef a short distance above Nicol's Town; shortly after
went around Morgan's Bluff, the rocky precipitous cliff fifteen or eigh-
teen feet in height that forms the northern end of the island. We
reached Lowe Sound about noon, then sailed for a long distance in

10 BAHAMAN TRIP

shallow water ; long sand-bars, some exposed and some under water,
and a few cays showing to seaward. On one were a number of terns
and a pair of oyster-catchers. Shot one of the latter ; as it fell in the
water a barracuda snatched at it, but was beaten off by one of the men
with a harpoon. There were a number of sharks about, some five or
six feet in length. Although our boat drew less than two feet, we
were stranded on a sand-bar until four o'clock. As a consequence it
was dark when we landed at Red Bays on the west side of the island.
As walking a,bout at night is impossible, we did not attempt to reach
the settlement, but slept in a palm-thatched chapel near by.

Tuesday, April 15. Walked a short distance back of the settle-
ment of Red Bays. The land was low, covered with palmettoes and
the brier tree (Terminalia, one of the plants which later proved to be
undescribed). Returning farther down the shore, first met with the
marly deposit so characteristic of the west side. It is a very fine de-
posit, and so soft that in some places I sank in halfway to my knees.
It extends back from high-water mark from an eighth to a quarter of
a mile. In some places along shore are mangrove swamps. There
were also some mangroves beyond high-water mark, but these are
small and apparently not thriving. About two miles below the little
chapel where we had spent the previous night, we again turned inland
across a savanna, a piece of level land covered with grass and rushes,
where we found a number of plants new to us. (See account of the
flora.) We then entered the pines and followed the path to an orange
grove in a coppet, known as Lewis Coppet, where the house that had
been vacated for us was situated. After some discussion three shil-
lings currency was agreed upon as the price of rent per day and two
shillings sterling to be paid the wife for the five days' cooking. This,
I may say in passing, was higher than the usual rates asked.

April 16. Mr. Keith left. Very warm, thermometer 82. The
walking is very rough, but not as bad as back of Nicol's Town. There
are many banana holes. The soft soil covers the depressions in the
rock for a long distance from the water.

April 17. Out shortly after dawn. Followed a path through the
coppet southwest, then on through grass and palmettoes to a small
patch of woods beyond. Near here there was an agave in bloom, the
first we had seen. The flower stalk rose about twenty-five feet in the
air and was about eight inches thick at a distance of four or five feet

BAHAMAN TRIP

from the ground. At the top the branches were thickly covered with
the golden blossoms. These are a great attraction to the birds, as we
have seen about it a number of the cocoanut birds, the Bahama fly-
catcher, and the Cape May warbler. The latter was seen sipping the
liquid in the flowers. Many insects and humming-birds are also flit-
ting about them. (It was here that we first saw the males of the bird
that proved to be an undescribed species of oriole later named after us
by Dr. Allen.) Shot off some of the flowers.

April 19. Up at sunrise; walked out to the agave again to take a
photograph. A short distance beyond the agave the palmettoes grew
fewer and fewer until at length there was nothing but the scattered
mangroves and here and there a clump of the prickly ''brier tree."
Farther yet nothing to be seen to the south and southwest but a level
plain dotted with small mangroves. To the north, a dark line of pines
was discernible. The rest of the day was spent in shooting and
skinning small birds. They were unusually numerous about the house.
For several days numbers of warblers arrived during the night, evi-
dently migratory birds on their way north. Having heard that fla-
mingoes, in native parlance " filimingoes," were plentiful some miles
farther down the coast, we decided to make a trip there after spending
five days at Lewis Coppet.

April 20. By 7.30 a.m. we were down at the boat. Got in the
small boat and were pushed out by the men halfway to the larger boat
that lay almost aground over a quarter of a mile from shore. About
12.30 we landed at a place called Cedar Coppet. The beach at high-
water mark was composed of the same fine chalky deposit as at Red
Bays, but it was whiter. The pines here come within a quarter of a
mile of the shore, only a mangrove flat lying between. We left Cedar
Coppet about two and sailed down the coast to what the men called
Loggerhead Creek, but which seems to be the Deep Creek of the maps.
It was about twenty-five miles south of Red Bays and was the first
creek of any size passed. It was a little over a quarter of a mile in
width at the mouth, but soon narrowed to about seventy-five yards.
Sailed in a short distance and anchored ; the tide falling, we were soon
aground. The shore was very soft, white above, blue gray beneath.
Back of the bordering mangroves was a bit of coppet on a slight eleva-
tion where we made a lean-to of palm leaves for the night.

April 21. Went up the creek about a mile in the small boat, then

12 BAHAMAN TRIP

left it and walked to a small pond. The land was level as fair as we
could see, with the exception of a small coppet here and there. A walk
of about three-quarters of an hour brought us near a pond, and in the
distance we could see a scarlet patch, our first sight of flamingoes.
We got within about three hundred yards when they flew away. There
were four of them. We walked to another lake where we saw five
more, and I got within two hundred yards when they too flew away,
looking very brilliant and showing the black of their wings. Walked
to a large mangrove in the lake and shot a switching-neck, a blue
crane, and two long-shanks. Walking in the lake was exceedingly
difficult, the water being about three inches deep and the mud, into
which I sank at every step, eighteen or more. The " tell-bill-willy "
was very common, its shrill notes being heard on every side.

Walked three or four miles farther to " Two-camp Lake," which was
so extensive we could see neither end of it. All the ground about here
is said to be covered with water in the "wet weather," in June and
July. These lakes are evidently low places from which the water
does not drain. It is strange that they contain so few mangroves,
only one or two large ones, while on the shores they are scattered
thickly.

April 22. Very strong northeast wind. Left our anchorage about
five o'clock and sailed to the mouth of the creek, where we lay until
midnight, when we got under way. Skinned birds on deck all the morn-
ing, and by three o'clock we anchored at Red Bays; spent the night
in a thatch hut about nine feet square and about six feet high in the
center; it had a door of thatch and no windows. The ground back of
the settlement is rough, but not as bad as at Nicol's Town.

Wednesday, April 23. Beat all day against wind and tide on the
return trip to Nicol's Town; wind blowing too hard to let us go around
Morgan's Bluff, so we landed at Money Cay in Lowe Sound and walked
to the village.

Friday, April 25. Walked to Cocoanut Point, left the path and
walked back into the woods, had hard work to get through the bush.
Saw a number of bromeliaceous plants and found that some of the
orchids were in bloom. In the spreading base of the leaves of one of
the former plants, a species of Tillandsia we had not found before, a
colony of ants had their home. The thorax was brownish red, the
abdomen black and pointed behind. They have stings, and when

BAH AM AN TRIP 13

running about after being disturbed carry their abdomens elevated,
as scorpions carry their tails. A number of pupae were noticed.

Tuesday, April 29. Visited the reef this morning. Inside the
main reef is a smaller reef of large coral, called ''red coral" here, the
same as the large specimens from Rose Island ; some were twelve inches
in diameter. These grew thickly together, and their flat spreading
branches were very handsome. In some places were masses of dead
coral that showed cavities and holes similar to those on shore. The
water inside the reef has a depth of two or three fathoms in the deep-
est parts, but outside gradually deepens. The main reef is about a
quarter of a mile from shore. From half to three-quarters of a mile
from shore the depth was found to be from fifteen to eighteen fathoms.
At ten fathoms the water was so clear that we found objects on the
bottom could be quite plainly seen through a water glass; at eleven to
twelve fathoms the bottom could be seen and objects of different color
easily distinguished. At fifteen fathoms only the dark and light patches
could be made out. We were taken to a hole that proved to be a sandy
space about a fathom deeper than the surrounding level. The bottom
inside the reef, and outside also, is said by the sailors to be as "hard
as rock." In some places fragments of coral lay on the bottom in the
same manner as I have seen them on the land. [Of this Andros reef.
Professor Alexander Agassiz, in his ''Reconnoissance of the Bahamas
and the Elevated Reefs of Cuba in the Steam Yacht Wild Duck,^^ Janu-
ary to April, 1893, says, "This reef, though narrow, is one of the finest
reefs I have seen, and the patches of corals and Gorgonias which flour-
ish between the reef and the shore are not surpassed in beauty by the
corals of any district known to me."]

We had spent six or seven weeks at Nicol's Town collecting plants
and birds, and towards the close of April prepared to go on to Conch
Sound, four or five miles farther south, where we had made a brief
visit a month before.

Wednesday, April 30. Mr. Keith came for us in his boat, which
was soon loaded with our baggage, while we decided to walk. The
path for the greater part of the way led through the pine-yard, and
here and there through a coppet, until we neared Conch Sound, when
it skirted the water until we reached Mr. Keith's house. The evening
was beautiful with a clear moonlight night. The "tell-bill-willy"
of the willet was almost constantly heard, and there were frequent

14 BAHAMAN TRIP

splashes in the creek, presumably made by the barracudas. Conch i
Sound is a shallow inlet or bay, filled with small cays and mangroves
and with a narrow winding channel. In places on the shore were
great piles of the shells of the conch {Strombus). It is quite an impor-
tant article of food on Andros, and it is in these shells that the pink
pearls are found for which the Bahamas are noted. While Conch
Sound was our headquarters, in company with Mr. Keith, we made a
trip to London and Stafford creeks, eight or ten miles down the coast,
stopping at Mastic Point on the way. An account of the trip follows.

Friday, May 2. Left Conch Sound about 9 a.m., wind nearly dead
ahead. A hard rain fell just before we reached Mastic Point. When
it ceased, we had our luggage carried to the house of Mrs. Bain, where
we were to stay. It is a small settlement of not more than ten or fifteen
houses built close together. It might well be called "Bainville" as
everybody was related to old Mrs. Bain, our hostess, who was the
patriarch of the settlement and whose mother and father were full-
blooded Congos. The houses were better and neater than at Nicol's
Town, some even had a little porch at one end, while others had
brightly painted woodwork. Several had flowers planted about the
dooryard. Mrs. Bain's house boasted a plastered floor, which was
kept neatly sanded, and some fine pieces of mahogany furniture.
There was, as usual, a table covered with an array of glassware and
crockery, such as goblets, decanters, wine-glasses, lamps, and lamp
chinmeys. It is curious that these articles are never seen in use,
but are evidently regarded purely as ornaments.

In the afternoon walked a short distance along a path through the
mangroves. In some places where the rocks were overflowed by the
tide, there were many sharp and jagged points. Later walked through
a field where I noted corals in the rock and picked up a fragment.
Beyond and farther inland was a slight elevation, on the south side of
which the rocks were abruptly worn off, their appearance showing that
the sea had once washed against the top of the hill.

Saturday, May 3. Started about seven o'clock in the morning, but
were no sooner under way than there was another hard shower, lasting
nearly an hour. The men poled a good part of the way, and we finally
reached a large mangrove which was about one hundred yards long
and fifteen to twenty wide. A pair of cormorants were perched on one
of the top branches, and in the mangrove were many Louisiana herons

At Conch Sound,

BAHAMAN TRIP 15

which the men called "blue cranes." There were a number of their
nests in the mangrove, a few of which contained greenish eggs. Shot
two of the birds. We sailed on to Wax Cut Cay where we had lunch.
Near the shore were many Strombus shells which varied considerably
in shape. In some the outer edge of the aperture was thin and ex-
tended, while in others it was much thicker and narrower. The men
called the thick-lipped ones "sambo" and the others the "broad-
lipped conchs." We found some intermediate specimens which were
thickened but otherwise like the broad-lipped, so perhaps the sambo
conch is only an old broad-lip.

From Wax Cut Cay we sailed to Stirrup Cay, then past Rat Cay
and Stump Cay, and between four and five o'clock poled up what we
supposed was London Creek, but after proceeding a short distance,
could see an opening through the cays seaward, showing we were not
in the creek at all. Went about four miles farther and then anchored
for the night, pitching our tent on low ground covered with under-
growth and with a few pines.

Sunday, May 4. Up early, troubled by sand flies. A short dis-
tance above our camp, we entered London Creek. The men say
that the land to the south of us is a cay without a name and runs up to
a lake at the head of London Creek, with which Stafford Creek is also
connected. Our general course was nearly southwest, with many
turnings. The creek in some places was narrow, but as a rule was
perhaps an eighth of a mile in width. The water was only two or
three feet deep, and the bottom muddy. In one place the whitish
slimy deposit was six feet three inches in depth. The course of the
creek was through the pine-yard; in many places the pines and pal-
mettoes came down to the water's edge, while in others it was bordered
with mangroves. We sailed and poled up about six miles until the
water became too shallow for us to proceed. We walked a short dis-
tance through some swash, and then through pines, hoping to reach a
lake that the men said was four or five miles off, but a heavy shower
threatening, we made our way back to the boat. After the rain stopped
we sailed back out the creek and down to Calabash Cay, at the mouth of
Stafford Creek. Spent the night in an unoccupied new house.

Monday, May 5. Found Stafford Creek entirely different in char-
acter from London Creek. The mouth was only about sixty feet
wide, the water six or seven feet deep, and the current swift. After

16 BAHAMAN TRIP

running west for a short distance, it turned abruptly to the south and
spread out into a lake-like expansion about half a mile wide. The
shores are of hard coral rock, and the pines come to the water's edge.
It narrowed again and finally became a deep narrow winding stream
that was fresh for the last mile or more of its course. At the extreme
head the water became shallow, the bottom rocky in places, and the
current quite strong, while numerous small streams trickled down the
steep banks, into which some had cut quite deep channels. This
narrow fresh part is called the ''lake." It was here that we col-
lected two species of bladderwort (Utricularia), the only fresh-water
plants that we saw in the entire island. We must have penetrated
twelve or thirteen miles into the interior, through pines and palmet-
toes all the way until we reached the head where we found swash
running back perhaps two or three miles. In some places the ground
was covered with mangroves for a long distance. Altogether the land-
scape looked very much like that of the west coast. We left the head
of the creek at about three o'clock and reentered the broad lake-like
part about five, camping at six on the north shore. Here the creek is
half a mile wide ; to the west it extends for two miles, and you can see
three or four miles to the east. It is bordered by pines with an under-
growth of palmettoes and brier tree.

Tuesday, May 6. Left Orange Landing, where we had camped,
a place with a house and a few orange trees, at about 8 a.m., reached
the mouth of the creek at about eleven. Here we saw Cacti for the
first time, the ''dildo" (Cereus Swartzii) which grew ten or twelve feet
in height. (We found them quite common at Deep Creek later.)
Yesterday the "doctor flies" were thicker than the men had ever seen
them. These troublesome flies attack the feet and legs especially and
inflict quite a painful wound. They troubled the men much oftener
than they did us. About half-past eleven we started on our return
trip and reached Conch Sound about half-past eight in the evening.

May 13. Our next undertaking was a walk across the island, from
Conch Sound to the west side. Apparently no one had ever gone that
way, for the men were unable to give any idea as to how long it would
take. On May 13, 8 a.m., we left Mr. Keith's with four men and pro-
visions for four days. Took a road through the pines, and in a half
hour reached the beginning of Johnson's Coppet, a rocky ridge rising in
the midst of the pines and covered with a hardwood growth. Here

BAHAMAN TRIP 17

the banana holes were more numerous than anywhere we have been.
Some of them were twenty-five feet in depth, but the majority only ten
to twenty feet deep and perhaps eight to ten feet in diameter. Others
again were shallower and only four or five feet across. Many of these
holes were lined with drooping fern fronds, the most conspicuous being
the maiden-hair (Adiantum tenerum) and the creeping Goniopteris,
the latter with fronds often two feet or more in length and rooting at the
tip. After a short rest entered the pines again. There was no path
beyond this, and we were obliged to cut our way through a dense growth
of brake, the fern commonly called "Maypole" here (Pteridium
caudatum). It was eight or nine feet in height. The ground was
exceedingly rough, and the weather warm. The course thus far has
been nearly southwest, and we have gone about four or five miles. The
pines here are mostly seven or eight inches in diameter three feet
from the ground. Some of them are larger and very tall, eighty feet
at least. Passed through a tract covered with what the men call "bed-
grass," a species of Andropogon; then through Maypole again. A few
palmettoes and cycads begin to appear. The rock is soft, but very
jagged. Measured a large pine here and found it to be 4 feet 9
inches in circumference, about seventy feet high. A short time after
this the pines began to be smaller and smaller, the palmettoes more
numerous, the rocks rougher and harder, and banana holes more fre-
quent. Now, at one o'clock in the afternoon, the pines are only three
or four inches in diameter; keeping on over a very rough piece of
ground, we finally see light through the pines. We reach the end of
them about two o'clock and look out over a level stretch of ground like
a prairie with a coppet on the horizon in the distance, the men say it
is mangrove. To the southwest a long point of pines with water
showing on this side. The level ground before us is very soft and damp,
a sheath-knife does not reach rock.

Wednesday, May 14. Left camp and started on the return trip
a little after nine, reached Mr. Keith's about half-past two.

Leaving Conch Sound May 22, we next went to Mastic Point,
four miles below, where we stayed at Mrs. Bain's, as before. Made
several trips to "the big mangrove" three miles below, where I secured
a man-of-war bird. Also got a number of other birds and plants not
collected before. A number of the days were more or less rainy. After
exploring the neighborhood, we secured the Herald again for a month's

18 BAHAMAN TRIP

cruise about the southern end of the island. Left Mastic Point
Tuesday, June 5. Our first stopping place was to be Fresh Creek,
fifteen miles or so down the coast. We passed Rat Cay, Stump Cay,
Saddleback Cays, Calabash Cay, and the wind getting light, we ran
into Stafford Creek and spent the night there.

Wednesday^ June 14. Beat all day against a southeast wind;
reached Fresh Creek in the evening. There is quite a large settlement
here, and the creek more nearly resembles a river than any we have
seen. At its mouth it is from fifty to seventy-five yards wide. The
banks are rocky on one side, eight or nine feet high, the water deep and
the current very swift. There is said to be no reef off the mouth of
this creek.

Thursday, June 5. In the afternoon walked along the "govern-
ment road," the only path at all worthy of the name that we had seen
on the entire island. It was about ten feet wide and connected Fresh
Creek with a settlement at Calabash Bay, about four miles northwest.
It runs through the coppet all the way, with here and there a field.
Found a number of plants not seen before, among them a pretty orchid.
The great white land crabs are unusually abundant about here. We
have had numbers of them about the house, and several times have
seen long processions of them walking near the edge of the bluff.
They frequently touched the ground with their claws and put them in
their mouths. Were they feeling for water ? The people here say
they go to the water to breed. I think there were more males than
females, and the latter had no eggs under the abdomen. These crabs
are common all over Andros and do great damage in the fields, eating
the melons, tomatoes, corn, and even the bananas, we were told.
Whole boat loads are taken to Nassau for sale in the market, and the
men said they were less plentiful than formerly ; on New Providence
they have been largely exterminated.

Friday, June 6. Went out to Goat Cay, that lies off and slightly
above the mouth of Fresh Creek. It is about three-quarters of a mile
long and perhaps fifty yards wide. Here were numbers of the bridled
tern {Sterna anathetus) which the men called '' egg birds." They flew
about us and followed us around in an inquisitive manner, uttering
among other notes a sharp "cur cur cur-r-r." They were so tame it
seemed a pity to shoot them. A flock of noddies {Anous stolidus)
flew back and forth on the seaward side of the cay but they were much

Fresh Creek.

BAHAMAN TRIP 19

more shy. The southern end of the cay was lower than the other,
being about six feet above the water. The surface was exceedingly
rough, the rock being worn away so as to leave vertical, more or less
cylindrical, columns varying from a few inches to over a foot in height.

Monday, June 9. Left about 7 a.m. and sailed up the creek
with a southeast wind. The course of the creek is first through the
coppet and then enters the pines. It widens once or twice and seven
or eight miles from its mouth spreads out into a lake-like expansion,
from three to five miles wide, containing small cays and mangroves.
We sailed to the head of navigation and then poled in the small boat
until we reached fresh water. Fresh Creek has two or three arms and
must penetrate twelve or fifteen miles into the interior. In very wet
seasons there is said to be water communication with Wide Opening
and the lakes on the west side. We spent a couple of weeks at this
settlement, finding the flora more varied than farther north. The
thermometer ranged between 82 and 84. The mosquitoes were very
troublesome for the first time.

June 15. Left Fresh Creek. Continuing our course south, we
passed High Cay and numerous smaller cays, finally reaching Bearing
Point (called Salvador Point on the charts) at the entrance to the
Northern Bight about five in the evening. Here the ridge seemed very
high, apparently the highest land on the island.

June 16. The next morning walked to a high point from which a
good view was obtained. It was about seventy or eighty feet high, and
one could look westward across the bight to the horizon, where a few
cays could be seen. The bight is three or four, in some places five,
miles from the northern shore to the cays that are scattered through
it. One of the largest of these is Wood Cay. The coppet at Bearing
Point has pines beyond it, and there are pines on some of the cays in
the bight. Going on board again, our next landing was at Cormorant
Cay, a small cay, perhaps a hundred yards long by forty or fifty wide.
Many cormorants were nesting here, and both old and young birds
were numerous. Some of the latter were still in the nests, but the
majority were able to walk. The little ones opened their mouths and
darted at my finger. Later we entered what the men called Loggerhead
Creek, one of the channels of the North Bight, and sailing through this
to the west side, made north and reached Purser Point at evening.
This point is near the southern side of what is called the Wide Opening*

20 BAHAMAN TRIP

a shallow bay or estuary two or three miles wide. We reached the
point about five o'clock. The sun was just setting back of the white
chalky beach which was fringed with an unbroken line of palmettoes ;
here and there was a mangrove clump, and beyond, nothing but water
and sky. A desolate scene with no sign of life anywhere.

Thursday, June 17. Walked back from Purser Point. The whole
shore deposit is much whiter than at Red Bays. In some places along
shore the white mud has been washed by the ripples so as to resemble
miniature cliffs. These were made very noticeable by the dense blue
shadows. In places the water had washed out the foraminifera and
left them here and there, in little masses along the water line. The
deposit is very soft. I ran a pole into it nine feet at a distance of a mile
and a half from shore, where, nevertheless, the water was only about
eighteen inches deep. Birds were plentiful. We saw many flamingoes,
some summer ducks, long-shanks, night-hawks, and herons. There
were many ponds where the water was about a foot and a half deep.
Returned to the boat and about three o'clock started up Wide Open-
ing, proceeded about a mile. Spent the night in the rowboat. The
mosquitoes innumerable.

Wednesday, June 18. Sailed up as far as possible, but we grounded
opposite the narrow channel marked on one of the charts as the River
Lees which connects Wide Opening with the large body of water
known as Turner Sound. It was an unusually dry season, and to our
disappointment we found it impossible to cross the bar at the mouth of
the channel. Our supply of water was exhausted, so we were obliged
to turn back and run down the coast again to what the men called
Cabbage Creek, which seems from the chart to be the other entrance
to Turner Sound. Anchored for the night about a mile and a half from
shore.

Thursday, June 19. Aground again about a mile and a quarter
from shore, took to the skiff, but were soon obliged to leave that also,
and finally to drag the boat over the flats to the channel that runs out
for about half a mile beyond the mouth of the creek. The creek itself
is about half a mile wide at first, but soon narrows to about seventy-five
yards, which width it retains for a long distance. The water in the
center is about ten feet deep and the sides so steep that it is difficult to
"set" up it. The shores are fringed with mangroves and palms. The
creek is very winding, and the palms always grow on the outside curve,

Cabbage Creek.

BAHAMAN TRIP 21

the mangroves on the inside. We went up four or five miles to
where the creek narrowed at the beginning of a large salt-water sound.
Here the tide overflowed the land, and we found the same white deposit
as on the shore. Redwing blackbirds and willets are common on the
swashes. Got out of the creek in the afternoon, and with a southwest
wind, started back for Loggerhead Creek. On the way passed a
sponging schooner of less than twenty tons, with eight boats in tow.
Two men to a boat and the cook made seventeen men at least on the
vessel.

Friday, June 20. Loggerhead Creek. In the morning the men
coming on shore with sponges for the crawls passed us, some of the
boats sculled by small boys from seven to ten years of age. Later
passed a sponge-crawl, built of stakes and resembling a weir. The
sponges are soaked in these inclosures for about five days, and are then
cleaned by holding them in the left hand and beating them with a
stick. They are then left on shore until ready to be taken home.
About noon the sky darkened to the north, and rain fell in torrents
for about an hour. We took the passage known as the Middle Bight,
passing many cays covered with palmettoes or with pines.

Saturday, June 21. Continued beating through Middle Bight,
finally reached our destination, Mangrove Cay, towards evening. This
is one of the largest settlements on the island, and it is here that the
magistrate of the island lives. Both he and Mr. Mathews, the repre-
sentative of the Episcopal Church on Andros, are white men. Spent
the night with Mr. Mathews. The next day walked back a short
distance. There are many small hills, the surface is rough, and there
are numerous banana holes.

June 23. Walked to Lisbon Creek, about four miles south of the
settlement. The creek forms the southern boundary of the Cay. The
road ran along the beach towards a long point. Near the head of the
point was a deep hole close to the shore in which Mr. Mathews said
the water was twelve or sixteen fathoms deep. The diameter of the
hole was perhaps a hundred feet. We crossed the point and walked
along the southern shore to a house at the mouth of the creek, where
we planned to stay several days. The coppet about here is largely log-
wood (HcBmatoxylon campechianum). It is said that when once
planted this will soon drive out all the native trees. Thermometer 86
in the shade.

22 BAHAMAN TRIP

Tuesday, June 24. Up at sunrise. Sailed up Lisbon Creek. Not
far above its mouth it widens and is no longer a creek, but a large lake-
like body of water, shallow and containing a number of large cays
covered with pines. On these large cays are quite a number of iguanas.
We saw several ; they run with almost incredible swiftness. Got two
small ones. They are said to occasionally attain a length of five or
six feet. The negroes hunt them with dogs which drive them in holes,
when they are easily caught. We were told they were formerly much
more abundant.

Wednesday, June 25. Walked back to "high coppet." The
land was high and contained many banana holes. The trees were
larger and taller than in the surrounding region. The largest was a
" horse-flesh" that measured five and a half feet in circumference,
four feet from the ground.

Thursday, June 26. Left Lisbon Creek, continuing our trip south,
stopped a short time at Kemp Sound. Saw a small gray owl and found
some new plants. Shore bordered with cocoanuts and houses for a
long distance, the high coppet behind them. No pines to be seen.
Slept in the rowboat.

Friday, June 27. Sailed to Henry Dames, a short distance below
Deep Creek. There are five houses here, surrounded with cocoanut-
palms. We found Mr. Dames hospitable and intelligent. He gave
us considerable information as to medicinal plants and the commercial
value of the different kinds of wood.

Saturday, June 28. Walked back into the coppet about four
miles over the "road" made by the surveyors. No pines in sight.
The ground was high and very rough, and banana-holes were numer-
ous. The place is known as the " high coppet, " and the trees are the
largest we have seen on the island. Horse-flesh (Lysiloma) and Ma-
deira (Swietenia mahogani), between two and three feet in diameter,
were common, and a number of "nakedwood" trees (Myrtus punc-
tatus) were over a foot through. We found many new plants. Igua-
nas are said to be plentiful farther back and to sell for sixpence a
pound. The negroes consider them a great delicacy. The rare blue
thrasher and the white-headed pigeon were abundant about here.

Sunday, June 29. Left Dames at six o'clock. Continued on our
way south. Beat against adverse winds all day, only got within two
or three miles of Grassy Creek. Scenery along shore much the same as
farther north, except the trees were slightly lower.

Coming to Anchor off the Dames' Place.

BAHAMAN TRIP 23

Monday, June 30. Sailed to Grassy Creek, which is within a few
miles of the southern end of the island. The creek is about seventy-
five yards in width. The men said that the swash began only a short
distance back. There was a party of turtlers on shore cutting up live
hawksbill turtles. A storm threatening, we did not sail up the creek any
great distance, but turned back for Smith's place. Thunderstorm very
heavy, but we were not in the worst of it. The sky to the north was
very black; saw a water-spout in that direction, the column slanted to
the westward. There was a dead nurse-shark on the shore, found it
to be 7 feet 11 inches in length, steel color below, darker above.
Near Smith's house was the largest ocean-hole I have seen. Smith
says he has found no bottom with a twenty-fathom rope. Surrounding
it, there is at low water nearly three-quarters of a mile of sand fiat, the
water being only about two feet deep at that distance from the shore.

Tuesday J July i. Walked back over surveyor's line. The trees
and bush had been cut down and left where they fell. The ground very
rough with large banana-holes. The swash was seen only a half mile
back so that the coppet is here much narrower than at Deep Creek.
Back to boat and waited until four o'clock for the tide to rise. Reached
Dames that night, just below Deep Creek. Arrived at Lisbon Creek
the evening of July 2.

Thursday, July 3. Packed and put specimens on schooner for
Nassau. We started about midnight, slept on deck. Very slight dew.
Reached Nassau evening of July 4, having been becalmed much of the
day. A few days later we took the steamer for home.

It may not be amiss to say something here of the inhabitants and in-
terests of Andros Island as we found them in 1890. All the settlements
were on the eastern ridge, with the exception of the small one at Red
Bays in the extreme northwest corner. The remainder of the western
coast was practically uninhabited and was visited only by the spongers.
The writer believes she enjoys the distinction of having been the first
white woman to set foot on this part of the island. On the east coast
Nicol's Town was the most northerly settlement, then came Mastic
Point, Stanniard Creek, Fresh Creek, Bearing Point, at the entrance to
the North Bight, then Mangrove Cay, Long Bay Cay, and Deep Creek.
At the time of our visit Nicol's Town, Fresh Creek, and Mangrove Cay
were the largest. All communication between the different settle-
ments was by water, as there were no roads, the government road at

24 BAHAMAN TRIP

Fresh Creek being the nearest approach to one. What the negroes
termed roads were only the roughest kind of footpaths leading back to
their fields. The ground was on the whole very much rougher than on
New Providence, and walking was often extremely difficult.

The people all seemed to be very poor. The majority of the men
were engaged in sponging, and the women in consequence did most of
the work in the fields. Some few owned their land, but the majority
cleared a piece of coppet and started a "field" wherever they fancied,
often four or five miles back in the bush. They raised corn, cassava,
sweet potatoes, bananas, and sometimes a few beans, pigeon-peas,
yams, and tomatoes, usually only a few plants of each, set out
at random. Their principal tool was the machete, with which they
cut down the weeds and bushes and turned up what little soil there was
in the cavities of the rock. A number of the gentlemen of Nassau had
interests and plantations on the island. There were extensive cocoa-
nut groves near Lowe Sound and Nicol's Town; sisal plantations at
Conch Sound, and at Kemp Sound just above Deep Creek, and a large
pineapple plantation near Long Bay Cay. These, with the two sponge
warehouses, one at Nicol's Town and one near Mangrove Cay, com-
prised the business interests of the island.

At the time we were on Andros there were only seven white people
on the island, and for weeks at a time we did not see a white face. We
found the negroes, without exception, courteous, hospitable, and
obliging. More than that, as far as we were concerned, we found
them perfectly honest. We frequently left our house open and un-
guarded all day, yet not a single article was ever missed.

We heard that many changes took place shortly after we left. A
large English company bought up a number of square miles and began
the cultivation of sisal on an extensive scale. After a few years' trial it
was found that the financial returns did not come up to the company's
expectations, and the work has since been entirely abandoned.

SUMMARY OF THE COLLECTIONS BROUGHT BACK

VERTEBRATES

Mammals.' — The only mammals on Andros were bats and rats. The bat was
Macrotus waterhousei, kindly determined by Dr. J. A. Allen. The rat was Mus

rattus.

BAHAMAN TRIP 25

Birds. — Two hundred and eighty-six specimens, representing twenty-seven
families and seventy-four species.

Reptiles. — Lizards, many specimens in alcohol. The collection was sent to
Dr. Cope for determination, but unfortunately the list returned by him was lost.
The following note is from one of his letters: "There is no new species, but the
Sphcerodactylus spectator is brought from the Bahamas for the first time, to my
knowledge." There were several iguana skins in the collection.

Snakes. — A number of specimens in alcohol, representing about eleven spe-
cies. The largest was a species of boa-constrictor.

Fishes. — About twelve species represented.

INVERTEBRATES

Mollusks. — Four or five hundred specimens, representing one hundred and
nine species. See list of shells named by Dr. Dall. There were also four or five
Ascidians and Cephalopods.

Crustaceans. — About 375 specimens, representing sixty-seven species. See
report by Dr. Rankin,

A number of centipedes, spiders, beetles, moths, and butterflies were also
collected.

Annelides. — Numerous specimens.

Echinoderms include, in addition to the starfishes (a list of which appear be-
low), many specimens of holothurians and sea-urchins.

Ccelenterates. — Numerous corals, including a number of alcyonoid corals.

Actinia. — Fourteen species. See report by Professor McMurrich. There
were also a number of hydroids and several medusae. Of one of the latter which
was sent to Dr. J. Walter Fewkes he wrote "the large black specimen I take to be
a Linerges. The color is brown yellow in nature. Linerges mercurius Haeckel
is figured in my Acalypha from Key West and Tortugas."

There was also a set of Bahaman sponges.

Plants. — About seven thousand specimens, representing five hundred and
seventy species. See report on Flora of New Providence and Andros.

A collection of woods.

Various geological specimens, including fossils, specimens illustrating erosion,
and samples of soil from various places.

With much of the material were full notes, for it was my husband's hope to
embody the results of the trip in a volume which would be a contribution to the
physical history, fauna, and flora of the Bahamas. The following pages show
all that it has been possible to do towards fulfilling this plan. The greater part
of the zoological material Dr. Northrop expected to work up himself. After his
death. Professor Rankin kindly consented to determine the crustaceans and Profes-
sor McMurrich to complete the work on the Actinia. It has unfortunately been
found impossible to have the rest of the collection worked up. It was stored for
several years, owing to the removal of the University, and many labels were lost
and misplaced.

26 BAHAMAN TRIP

The entire collection is at Columbia with the exception of some of the dupli-
cates. A set of the plants is in the University Herbarium in Bronx Park.

LIST OF STARFISHES 1 l^ J

Determined by John I. Northrop

Linckia Guildingii Gr. Collected] at [Nassau and [also on Andros.

Echinaster lentus ? Nassau.

Ophiocoma echinata Ag. Common in pools and under stones at
low water.

Ophiocoma pumila Liitken. Nassau. Twenty-one specimens. In
one the upper arm-spines are flattened vertically, and the tip is
indented or sometimes toothed. A few spines are much
broader at base and prolonged backward in a point over the
upper arm-plates, making them appear arrow-shaped.

Ophiocoma Riisei Liitken. One specimen from near Nassau.

Ophiura brevicauda Lyman. One specimen from near Nassau,
under stones.

Ophiura cinerea Lyman. Nassau. The seventh joint [of arm has
ten spines instead of eight. '

Ophiura appressa Say. Nassau. Four specimens.

Ophiomyxa flaccida Liitken. Nassau. Single specimen from near
Quarantine Station.

Ophiothrix (Erstedii Liitken. One specimen. Lyman states that
this is the most abundant species of Florida and the West
Indies. It did not seem to be at all common in the Bahamas.

Ophiothrix Suensonii Liitken ? Five specimens found on sponges in
the sea-garden near Nassau, seem to have but three or four
spines. ^

^ Owing to their incompleteness the critical notes have been omitted.

NOTES ON THE GEOLOGY OF THE BAHAMAS^

John I. Northrop

(Abstract)

By far the greater number of the islands known as the Bahamas
are situated on the Great and the Little Bahama Banks. The former
occupies a large area southeast of Florida and north of Cuba, and is
penetrated in the northern part in a remarkable manner by a tongue
of the ocean in which the water is over seven hundred fathoms in
depth. Little Bahama Bank lies north of the Great one and is much
smaller.

The islands lie, almost without exception, on the northern and
eastern edges of the banks, the main exception being the island of
Great Bahama, which is situated on the southern edge of Little Bahama
Bank. Southeast of the Great Bank is a small one, on which are
Crooked and Acklin islands, the former on the northern, the latter on
the eastern edge.

The Caicos Bank lies still farther southeast and contains on its
northern border a number of small islands. The rest of the group —
viz. the islands of San Salvador, Rum Cay, Atwood's Cay, Mariguana,
and Little and Great Inagua — are differently situated from those pre-
viously mentioned, rising almost abruptly from the deep ocean.

The time spent in the Bahamas was a little more than six months,
two of which were passed in New Providence and the rest in Andros.
The former is well known as the seat of government and as a health
resort. It lies on the northern edge of a portion of the Great Bahama
Bank, and is about twenty miles long and seven in width. A ridge
runs along the northern side, the highest point of which is 109 feet
above sea-level, and is occupied by an old fort — Fort Fincastle. Here
a fine view of the general features of the island may be obtained. To
the south stretches a low level country dotted here and there with

^ Trans. N. Y. Acad. Sciences, October 13, 1890, pp. 4-22.
27

28 NOTES ON THE GEOLOGY OF THE BAHAMAS

cocoanut groves, and in the distance is seen a lower ridge, known as the
Blue Hills. To the north we look over the harbor, half a mile in width,
to Hog Island, a cay about four miles long, perhaps half a mile wide,
and about twenty feet high. The eastern end is separated by a narrow
piece of water, known as- the ''Sea Garden,'^ from Athol or Long
Island, locally known as Quarantine Station, a cay about two miles
long and but a quarter mile wide, and continued eastward by narrow
shoals. West of Hog Island is the channel, or, as it is called, "the
bar," on the other side of which a number of small cays continue the
same general direction as the longer axis of the island. Seaward of
Hog Island, and separated from it by about one mile of water, is Salt
Cay, an island about two miles in length and a quarter in width. Far-
ther to the north and outside of Quarantine Station is Rose Island,
another long narrow cay. Outside of all lies the reef. These facts
have been given with perhaps more detail than is interesting, but which,
nevertheless, is necessary, as it has a bearing upon what follows.

Andros Island is the largest of the Bahama group, being over
ninety miles long and between forty and fifty miles wide. The northern
portion is separated from the southern by a broad shallow sheet of
water that contains many cays, large and small, and the passages
through from the eastern to the western coast are known as bights.
It is interesting to note that Andros is not only the largest of the Ba-
hama islands, but the largest coral island in the world, its northern
half alone having a superficial area of over 1200 square miles, while
the area of the whole island, so called, is over 1900. It Hes near the
eastern edge of the bank that faces the tongue of ocean referred to
above.

The eastern side of Andros is occupied by a ridge, of which the
highest point that I visited, and I think the highest on the island, is
marked on the chart as Salvador Point, locally known as Bearing
Point. Its height I had no means of determining, but I judged it to be
about 100 feet. At the extreme southern end, as at Grassy Creek,
this ridge is absent and the land is low.

When one has crossed the ridge, which is covered with angiosper-
mous trees, he enters a large forest of Pinus hahamensis which occu-
pies the greater portion of the interior of the island. The land is here
almost level, and in some places, as near Loggerhead Creek and Wide
Opening, there are extensive lakes of fresh water. Upon passing

NOTES ON THE GEOLOGY OF THE BAHAMAS 29

through the pines and reaching the west side, one comes to the most
remarkable feature of the island. Here one sees spread out before
him, as far as the eye can reach, a low swampy country covered
with small mangroves, Conocarpus and Avicennia, bounded on one
side by the water, and sometimes, in the distance, by a dark line of
pines. The pines, however, frequently jut out in points that approach
quite close to the water's edge. This level swampy land is locally
known by the very appropriate name of " swash." To the west of the
land stretches the Great Bahama Bank for a distance of from fifty to
seventy miles, and the slope of the bottom is so exceedingly slight that
at the distance of seventy miles from shore the water is but three or
four fathoms deep. The bank then plunges suddenly into the ocean
beyond.

There are many creeks on the island, and the water in all, at a
distance of ten or fifteen miles from the mouth, is drinkable. Many
little streams of fresh water flow into these creeks, thus partially drain-
ing the immense area of swash.

The creeks are generally narrow and winding, and by wearing
away the land on the convex side of the curves change the character
of the surface of the country. This was most plainly seen up Cabbage
Creek, near Wide Opening, on the west side. Here, as the creek wore
its way into the land, it was followed on the concave side by a growth
of small mangroves, while its convex side was fringed with palmettoes.
As the creek, in winding, changed its course, the palmettoes and man-
groves changed sides, as it were, the former always on the outside of
the curve, thus making quite a striking alteration in the appearance of
the landscape.

The surface on the western side of Andros is composed of an
exceedingly fine, almost impalpable, calcareous coral "mud," that also
forms the bottom of the shallow water that covers the bank. As we go
back from the water's edge this deposit becomes harder and harder,
until finally it is cemented into a hard, very fine-grained rock that is
very different in appearance from the rock of the eastern coast. The
latter is ^olian and varies in texture. In some places it is quite coarse
and in others fine, but always composed of rounded grains of coral
sand or comminuted shells. In some places, as at Conch Rock, near
Conch Sound, the rock has been so altered as to resemble a dark gray
crystalline limestone. In one place only did I see the contact between

30 NOTES ON THE GEOLOGY OF THE BAHAMAS

the JEoliaii formation and the true coral rock. This was on Andros
near the entrance to Fresh Creek, where a vertical face about ten feet
in height was exposed. The lower four feet were composed of frag-
ments of coral stems, mostly a madrepore, probably Madreporus cer-
vicornus, but so eroded as to make identification impossible. Above
this lay the ^Eolian rock, the contact being sharp and distinct, and the
contrast between the laminated appearance of the one and the irregu-
lar surface of the other being very striking. While no masses of corals
were to be seen, that were undoubtedly standing as they grew, the
fragments, some of them about a foot in length, were crowded to-
gether and presented the same appearance as I had seen in places in
the reef at Nicol's Town, where the branches of M. cervicornus had
been broken off and lay in a heap on the bottom. The striking con-
trast between the ^Eolian rock above and that in which the corals
were showed plainly that the two rocks were formed under different
circumstances, and that the lower stratum had been formed under
water, for it showed none of the characteristics of beach or wind-drifted
deposits. Although this is the only place where elevated corals were
found in profusion, in many other localities, as at Nicol's Town and
Mastic Point, Andros, a few fossil corals were found embedded in
the rock and evidently in place. The most common coral thus found
was a Maeandrina, probably the same as that now living on the
shore near low-water mark — a circular, flat species.

In Nassau excellent sections of the rock can be seen in an old
quarry, the *' Queen's Staircase," and in the present quarry. In the
first of these places the section is, I believe, ninety feet high, and the
rock is entirely ^olian. The most common fossil is a species of Stro-
phia that I suspect will prove different from those now living on the
island.^ A few helicoid shells were collected, and parts of a crab were
noted, but only one or two marine shells, a Fissurella, and fragments
of a lamellibranch were obtained. These facts all indicate the ^Eolian
origin of the deposit.

* My thanks are due to Professor Dall, who has kindly examined my Strophias,
and who is inclined to provisionally refer some of the fossil forms to S. maritima Pfr.,
and states that others seem to connect maritima with forms like Glans Kust., which I
found living on Andros.

NOTES ON THE GEOLOGY OF THE BAHAMAS 31

Surface Deposits

There is practically no soil either on New Providence or Andros,
except the little that collects in the caves and holes in the rock. From
the caves a large quantity of cave earth has been taken and sold as a
fertilizer, with what result I do not know. I was told that guano had
been found on Andros, but I saw none, The earths I collected I have
treated with hydrochloric acid, and the results indicate a great varia-
tion in composition. In some, most of the material is soluble with
strong effervescence, but in others the residual, I suspect, will prove
greater than the soluble portion. The residue seems to be composed of
organic material, with some mineral constituents that contain an appre-
ciable quantity of iron. Dr. Waller, of the School of Mines, Columbia
College, has kindly offered to analyze my material, and we hope from
the results to draw some conclusions as to its origin. At present it
seems as if the earthy deposits are composed of the insoluble portions
of the coral rock mixed with vegetable particles.

The absence of any large quantity of fallen leaves is very noticeable.
In the *' Pine-yard, '^ as the pine woods are locally called, we did not
see a single fallen pine, but the depressions in the ground showed
where one had formerly stood, and a long shallow trough plainly indi-
cated where the fallen trunk had lain. Desiccation and decay take
place with astonishing rapidity, and I believe that the deposits found
in the caves are greatly added to by the comminuted particles of dried
vegetable matter, and the residue of the decomposed portions, both of
which would be washed in by rain-water.

On the "swash" on the west coast of Andros a peculiar ash-like
deposit was noticed that covered small circular areas. This, on exam-
ination, proved to be composed of soft elastic lumps coated with cal-
careous material. These lumps under the microscope were seen to be
composed of one of the small, probably fresh-water, algae that had
grown in the pools of water, and which had been left when the pools
had dried.

Erosion

One of the most striking geological changes taking place in the
Bahamas is the erosion of both the surface and the shore line of the
islands.

32 NOTES ON THE GEOLOGY OF THE BAHAMAS

West of Nassau, on the beach, are many places where the waves
have cut narrow passages into the rocks and ground the fragments
into sand. There are also a number of ''pot-holes" ; these were circu-
lar in horizontal sections, and the bottom and sides were smooth. They
contained the worm coral heads that had evidently been the means of
making the holes.

On the beach of the cays north of Nassau great blocks of coral rock
are seen that have been dislodged by the waves, and in some places
fresh fractures show where large fragments have been broken off. On
the south side of the cays the erosion is different. Here the water is
quiet and eats slowly under the rock, so that a projecting ledge is
formed that marks the height of the tide.

At some places on Andros, as at Fresh Creek and Nicol's Town,
the shore is undermined, and great slabs in places have fallen, thus
making large cracks. Similar cracks are found inland higher above
high-water mark, and have evidently been formed in the same manner
as those on the shore. Where the edges of the crack were in contact
they had been firmly cemented together, and at intervals along the line
of the crack were numerous large holes that had evidently been worn
out by the action of rain-water running over the edge. Leading into
these holes were channels almost as perfectly formed and rounded as a
tin gutter.

The erosion of the surface will impress every one that visits the
Bahamas. Sharp jagged points project so as to make walking tire-
some and annoying. Although there is some sand near the beach west
of the Barracks at Nassau, there are no great moving masses such as
Heilprin describes as being found in Bermuda.^

In some places, as on Goat Cay, near Fresh Creek, Andros, the
surface is eroded in a peculiar and striking manner. There the rock is
worn so as to leave vertical cylindrical masses two or three feet high,
some connected below or halfway up with each other, others separate.
Their sides and tops are pitted and eroded, so they have evidently been
affected by atmospheric agencies. On the ocean side of this cay the
erosion from the action of the surf was so exceedingly rapid that the
rock remained a light yellowish color instead of the dull or dark gray
it commonly assumes. The form the rock here presented was even
more striking than the cylindrical masses described above. It was

^ "The Bermuda Islands," p. 31.

Goat Cay, Andros.

NOTES ON THE GEOLOGY OF THE BAHAMAS 33

worn into innumerable peaks and pinnacles like a miniature mountain
range, the points and edges of the pyramidal projections being sharp
and clean. As we walked back from the edge of the cay, every grada-
tion could be found between the miniature peaks and the cylindrical
masses higher up. And I believe that the latter are what remains
after the edges of the little peaks and pyramids have been slowly worn
away by the action of atmospheric agencies ; for we have only to round
off the points and deepen the connecting ridges to produce the vertical
cylindrical masses. But in order that this could have occurred, it will
be necessary to suppose that the island has been elevated, and this, as
I hope to show, I believe has been the case.

Outliers

In many places near Andros, as at Mastic Point and Golden Cay,
there are cays separated from the island by water but a few feet in
depth, and in some places these cays make a prolongation of a point
with which I believe they were formerly connected, and have been cut
off, not by subsidence, but by erosion. There are other cays, as those
near Nassau, that, as I hope to be able to show, owe their origin to
another source.

Caves

One of "the sights'' at Nassau are "The Caves," about seven
miles west of the city. One of these is an irregular opening in the north
side of a hill that faces the sea. The floor is considerably above high-
water mark. In the back of the first chamber is a small opening through
which can be seen a deeper chamber, in the bottom of which is water.
This chamber is said to connect with the cave on the opposite side of
the same hill. This second cave is a long chamber about fifteen or
twenty feet in height, and the roof contains holes through which the
roots of trees pass and fasten into the floor below. The side wall of the
cave for a distance of about four feet from the ground projected about
six inches beyond the upper part of the wall, thus forming a shelf that
was quite level and ran the entire length of the chamber, a distance of
perhaps one hundred feet. I could only explain the formation of this
shelf by supposing that it represented the contact between two depos-
its, and that the upper had yielded more rapidly than the lower to ero-
sion.

34 NOTES ON THE GEOLOGY OF THE BAHAMAS

On Andros I saw a number of caves near the northern part of the
island. All were in the sides of the elevated portions. The openings
in some, as one near Nicol's Town, were small, barely as large as an
ordinary door. Others were simply excavations in the side of the
hills.

In some places, as near Nicol's Town and Mastic Point, small
caves were found, twenty to forty feet above high- water mark; and the
low vertical cliffs in which they were, indicated the existence of an old
shore line, for the rocks were undermined in the same manner as those
we now see on the present shore.

In some of these caves Indian relics have been found, and also
human bones, and I obtained a portion of a human humerus from one
on Andros.

Ocean Holes

Near the caves at Nicol's Town was a hole known as the " Ocean
Hole." It was about one hundred feet in diameter and perhaps forty
feet in depth, and contained a pool of brackish water. In one place
the wall had been excavated so as to form a small cave, while the rest
was quite steep and covered with trees and large blocks of coral rock.
In one place was a very marked unconformability in the rock, the
seeming strata below lying at an angle of about 30 degrees, while
above the layers were horizontal. This was probably only an unusual
arrangement of the ^Eolian formation; but in the Queen's Staircase,
where an unusually fine section is exposed, the layers lie at different
angles, but are wedged in between each other, as it were, and no such
sharp unconformability as that above described is to be seen.

The name ''ocean hole" is also applied by the natives of Andros
to deep holes under the water. Some of these are remarkable. The
first that I saw was near Mangrove Cay. Here, close to the shore, was
a nearly circular hole at least 100 feet in diameter, and in which the
water was said to be over 18 fathoms (104 feet) in depth. I did not
have an opportunity of sounding it, but the dark blue color of the water
told its own story.

While sailing up Fresh Creek, Andros, we came to another ocean
hole, which I examined. It was about ten miles from the mouth of the
creek, close to the northern bank, and about one hundred feet in diam-
eter. From the shore the water for a distance of about fifteen feet

Ocean Hole, Nicol's Town.

NOTES ON THE GEOLOGY OF THE BAHAMAS 35

was two feet in depth, and then suddenly deepened to eighteen feet over
a projecting ledge. Sounding across the hole did not show a greater
depth. The bottom of the hole was of soft coral mud. The bottom
of the creek surrounding the hole was covered with about two feet of
water, and in some places gradually sloped into the hole. Still farther
up the creek another hole was seen, but was not examined. The most
remarkable ocean hole that I saw was one near Grassy Creek, near
the southern end of the east side of Andros. The diameter was about
one hundred and fifty feet, and the shore itself formed one edge of the
hole. The sides were of sand at its angle of repose for a depth of about
six or seven feet below, and resting on an overhanging ledge of rock.
Where the tide had fallen it left the hole surrounded by at least a quar-
ter mile of sand flat on the ocean side, while, as stated above, the shore
formed the rest of the boundary. This hole I sounded with all the
line I possessed, but at twenty fathoms the weight was cut off, and I was
unable to obtain another to continue the sounding. The captain of
our boat, a reliable man, told me that in the "Pine-yard" was a hole
similar to this one, in which he had found no bottom with a line over
thirty fathoms in length. Before attempting to explain the formation
of these ocean holes it will be well to describe what the natives call

Boiling Holes

The first of these I was shown on Andros in a small creek that runs
into Conch Sound. The top of the hole was about a foot under water
at low tide, and close to the mangroves that formed the side of the creek.
It was about seven feet in length and about two or three wide. Below
the diameter increased, forming an overhanging ledge. When the tide
was low in the creek, but rising outside, the clear sea-water could be
distinctly seen ascending, thus producing the same appearance as
that presented by a mixture of sulphuric acid and water. Suspended
particles could also be seen rising.

While sailing past Rat Cay, near Mastic Point, another "boiling
hole" was seen that was apparently about ten feet in diameter, and
from a distance we could see a perceptible "boil" on the surface that
was undoubtedly caused by the rising water. Our captain said that
when the tide was falling, the water in the hole went "down and
round" — which statement I believe, as the water was rising with some

36 NOTES ON THE GEOLOGY OF THE BAHAMAS

force, and probably ran out again with sufficient rapidity to cause
a small whirlpool. In another boiling hole near Mangrove Cay the
water was seen ascending.

These facts prove not only that an underground connection exists be-
tween these holes and the ocean, but that the connection is an open one,
so that the water can flow freely through it, and thus the pressure result-
ing from the passing tidal wave is shown before the tide commences
to rise on the shore. The ocean holes, I believe, can be explained
by supposing them to be old boiling holes in which the connection has
been stopped up, and their greater size caused by the falling-in of the
ledge on the edge, which would aid in the stoppage. I regret that I
have no facts to offer on the depth of the boiling holes, for the only one
I stopped to examine was at Conch Sound, and this one ran under the
ledge, so that its depth could not be determined. The ocean hole at
Nicols Town, described above, is also, I believe, an old ocean hole
now elevated.

Banana Holes

These are holes found mainly on the elevated parts of the land on
both New Providence and Andros. They contain a quantity of earth
in the bottom, in which the natives plant their bananas, hence the
name.

In shape and dimensions these holes vary greatly. Some are cylin-
drical, about four feet in diameter, but at the same time twenty feet or
more in depth. Others are from ten to twenty-five feet across, and
some even larger, and often of an irregular shape and much longer
than wide.

The walls are often excavated below, so that the side becomes an
overhanging ledge and forms a small cave. In some a cave begins at
the side of the hole and runs backward. It is hence hard to draw the
line between these holes and the caves. There are also holes that are
not called banana holes, but which may be here described, as they dif-
fer only in shape. In some the opening is barely large enough to allow
a man to pass. One such I descended, and found that below it was over
five feet in diameter and cylindrical. The top had been excavated so
as to form a domed roof.

Other holes were connected by a horizontal passage through which
I could crawl from one to the other. One of these I saw near Conch

NOTES ON THE GEOLOGY OF THE BAHAMAS 37

Sound, where the passage ran from the bottom of one hole to the side
of the other, which was much deeper.

Near by I saw two shallow holes that were connected by a horizon-
tal passage, so that they resembled a large tube bent up at each end.
It is not unusual to find openings in the ground, barely large enough
to admit an ordinary pail, and sometimes much smaller. These are
simply openings in the roof of a cave or hole of unknown dimensions,
and frequently in the bottom is a quantity of fresh water that is used by
the people.

The subject of banana holes has been briefly discussed by Dr.
C. S. Dolley,^ who accounts for their formation by "the action of
decaying vegetable matter, that undergoes fermentative changes by
the products of which the soft calcareous rock is dissolved and leaches
away." There is no doubt that the rock is in many places eroded in
this manner, as the small saucer-shaped depressions so common on
the surface, and each often containing leaves and water, plainly testify.
But I doubt if this agent alone would cause the deep vertical cylindri-
cal holes, or those in which the sides recede into caves or the horizontal
passages. And if the holes were formed in the manner described by
Dr. Dolley, should we not find them in the low-level land as well as
on the ridges ? But, as stated before, the holes are found in far greater
number on the ridges, and in places where the surface is such as to
indicate that formerly the erosion from the waves was very great. If
we could have a good view of the proper bottom of a banana hole, we
might be able to account for their formation; but, unfortunately, the
bottom is always filled with a deposit of earth or blocks of coral rock,
and generally covered with vegetation.

It is not improbable that the deep cylindrical ones were formed in
the same manner as pot-holes. And others might have originated in
the same way as the spouting holes, where the waves undermine the
shore and afterward break an opening in the rock above. Should
some of the boiling holes, described above, become elevated and their
bottoms filled up with fallen blocks of coral rock and deposits of earth,
they would form banana holes. The holes in the cracks at Fresh Creek
can be easily explained, but if all banana holes were formed in this
manner, we should find them in a line with others, which I was told was
the case, but I was never able to satisfy myself that it was so. The

^ Proc. Acad. Nat. Sci. Philadelphia, 1889, p. 132.

38 NOTES ON THE GEOLOGY OF THE BAHAMAS

caves are often the result of the former action of the sea, and some of
them have probably been washed out by rain-water; but in either case,
should a portion of the roof fall in, it would make a banana hole if
small, or, if large, an ocean hole like the one near Nicol's Town. The
horizontal passages are evidently washed out by water, but whether by
the sea or rain-water, I do not know, but I believe either might have
accomplished it. It is difficult to understand how underground chan-
nels could be formed under water, yet the boiling holes prove that such
exist; but there is no means of determining whether they were formed
under the present circumstances or at some previous period when the
land might have been elevated.

I was told that holes were as common under the water as they were
on the land, but did not myself observe this to be the case; but then
my opportunities for observation in this direction were limited. I infer
from the facts I have given that banana holes and caves pass gradually
into each other, and that they have been formed by the action of the
sea-water and afterwards modified by the action of rain-water, aided
by the products of the decomposing vegetable material and in some
cases by the falling-in of the roofs of the caves.

Effects of Vegetation on the Surface '

One of the facts that I noticed shortly after my arrival in the Baha-
mas was the occurrence of great numbers of blocks of coral rock scat-
tered irregularly over the ground, and I first thought that they were
the result of the excessive erosion that I saw taking place around me.
But on some of the cays — as on Goat Cay, described above — where
the erosion was most rapid, there were no loose blocks, and if these
had been formed by erosion alone, it was there that we should have
found them most numerous.

I had often noticed the gnarled and stunted appearance of the
bushes and trees that grew near the shore, and where there was evi-
dently a severe struggle between the sea on the one hand and the
plants on the other.

At Quarantine Station I was shown a small bush, Rhacicallis
rupestris, and was told that it was over twenty-two years old, by a man
who said he could remember the plant as ''long as he could remember
anything." I cut away the rock surrounding the bush, and found that

NOTES ON THE GEOLOGY OF THE BAHAMAS 39

its roots proceeded downward a few inches through the soHd rock. At
the surface the trunk expanded so as to form a projecting mass that
rested on the rock. The bush was only about eighteen inches in height
and evidently stunted by lack of nourishment. I then pulled up a
number of shrubs in the vicinity, and found that their roots ran under
the hard crust that formed the surface of the rock. Further observa-
tion showed that in many places the crust had been lifted and broken by
the growth of the roots, and then trees were noticed with the base of
their trunks surrounded by slabs of rocks that leaned against them.
Finally, when the rock becomes eroded, the roots of trees penetrate the
holes and crevices, and by their growth crack off large fragments that
subsequent erosion forms into boulders. And these boulders are found
most abundant where the trees are the largest, and hence where the
action of their roots is most vigorous. Hence we may infer that these
blocks are formed by erosion and the growth of the roots of the shrubs
and trees.

While at Nassau I noticed on the shore in some places — as near
Dix Point — vertical masses of rock that ran in an irregularly curving
or straight line on the surface. The projections were sometimes about
one foot high and two to four inches in width, and containing on their
upper edge a number of holes, about half an inch in diameter, that were
often shallow and sometimes mere pits, but quite regularly distributed
in a single line. These little ridges could sometimes be traced for a
distance of fifteen or twenty feet, and had evidently been left standing
while the surrounding rock had been eroded. For a long time I was
unable to explain their formation. Sometimes the ridges intersected
and formed small or large triangles, as the case might be. When I
visited the south side of New Providence, I saw growing in the water,
in the calcareous mud that there forms the bottom, the black mangrove
or salt bush (Avicennia nitida), and radiating from it, projecting about
five or six inches above the ground, were small vertical shoots about
three or four inches apart and looking very much like the teeth of a
very long wooden rake. These shoots, I afterward found, come from
the long sucker-like roots of the Avicennia and also of the button-
wood (Conocarpus erectus). Later, on the west side of Andros, I found
these plants growing near the water and also higher up on the beach,
which here was a very fine calcareous deposit. This deposit had been
slightly raised by the growth of the shoots, and higher up on the shore,

40 NOTES ON THE GEOLOGY OF THE BAHAMAS

where in contact with the shoots, it had hardened into rock. This,
I beheve, explains the ridges described above, for the roots are fre-
quently a foot below the surface, and the action of fresh water follow-
ing down the shoots, and probably, also, the action of the juices of
the plant, have slightly dissolved the calcareous material and then
cemented it into a rock harder than the surrounding deposit. The
latter, when erosion commences, wears away and leaves the harder
ridge standing. The manner in which the ridges are formed, and the
holes in the tops of them, all strengthen this conclusion. The trian-
gular spaces enclosed by the ridges are almost identical in appearance
with the triangular spaces marked off by the sucker-like roots that
cross each other in all directions.

Rhizomorphs

While visiting the quarry at Nassau my attention was drawn to
some cylindrical masses of coral rock that apparently hung root-like
over the edges of the quarry and were about four feet in length. They
were, however, cemented to the wall. I broke one off, and on examin-
ing it, found that the particles of which it was composed were arranged
in a concentric manner about a central axis. On the way back from
the quarry I pulled up a small shrub, and found its roots penetrating
the rock, which had been eroded so as to leave cylinders surrounding
them. Dr. Dolley ^ has called attention to these cylindrical and tubu-
lar forms, and has explained their formation by supposing that they
represent the "ramifications of a now exterminated flora," and also
that '' the juices of the roots, acting on the sand immediately surround-
ing them, formed a compact layer. Through erosion and subsidence
the vegetation was afterward exterminated, the looser particles of drift
rock worn away, and the surface left covered by myriads of tubes of
all sizes, formerly occupied by plant roots and rootlets." Later, on
Spruce Cay, near Nassau, and at many other places, I found a number
of these cylindrical projections, some of which contained a small hole
in the center, which in others was filled with calcareous material. I
believe Dr. Dolley's theory of the formation of these cylindrical and
tubular projections is, in the main, the correct one; for I collected
specimens with the roots still in them, and the concentric arrangement

^ Loc. cit.f p. 131.

Spruce Cay, New Providence, showing Erosion.

NOTES ON THE GEOLOGY OF THE BAHAMAS 41

shown on cross section shows very plainly that the solvent action of the
water, following the path of the root, had been from within outward.
As these masses are conspicuous and characteristic, I would suggest
that they be called rhizomorphs ; and this name would also cover the
tubular masses that have evidently been formed in the same manner,
and which I have found in the ferruginous clays of New Jersey and
elsewhere. The action of the water in the latter case has been on the
iron contained in the clay.

Dr. Dolley is inclined to think that the rhizomorphs were formed
while the sand was in a loose state; and while I do not deny that this
might be possible, I believe that all that I saw had been formed by the
roots penetrating the already hardened rock; and I furthermore found
no rhizomorphs being formed by the plants growing in the loose sand.

Dr. Dolley also states^ that the "small islands exhibiting these
peculiar formations [rhizomorphs] are indications, therefore, of
erosion and subsidence." I admit the erosion ; but as the rhizomorphs
are found twenty and thirty feet above the sea-level, I think that
subsidence has taken no part in their formation.

These rhizomorphs are not to be confounded with the cylindrical
masses described on Goat Cay, for the latter were vertical and did not
exhibit the concentric arrangement of the particles so plainly shown
in the rhizomorphs, which are found at all possible angles. And if we
suppose the masses on Goat Cay to be due to the same process that
formed the rhizomorphs, we must suppose that the roots that produced
them grew vertically downward, which is extremely improbable. It
may be well to state that a few rhizomorphs were found on Goat Cay,
and the difference between them and the cylindrical masses was very
marked.

It might be asked, if these rhizomorphs have been formed in the
manner described, why is it that we do not find them everywhere
on the islands ? And in answer to this objection it may be said that
roots of most of the trees spread out over the ground or slightly under
the crust, but do not penetrate the rock; and that the roots of the
smaller plants are those that have formed the rhizomorphs, and that
these latter show only where there is but little vegetation, and where
the erosion is active and not helped by the roots of trees breaking up
the surface into blocks as described above.

^ Loc. cit., p. 132,

42 NOTES ON THE GEOLOGY OF THE BAHAMAS

Evidence of Subsidence and Elevation

The facts bearing on the question of subsidence and elevation
have already been given, but it may be worth while to briefly review
them and give my inferences.

The caves and old beaches now above the sea plainly show where
the level of the water formerly was. The section at Fresh Creek
proves, as do the other cases of elevated corals, that the island has been
elevated. As the formation on top of the corals at Fresh Creek is
^olian, it follows that it could have been deposited only above water;
and as the caves and old beaches are at least thirty feet above the sea-
level and in the ^Eolian formation, they could not have been formed
until the islands had subsided. Hence we are justified in assuming
that at some previous time in their history the islands were at about the
same level as now. Then followed a period of subsidence of at least
thirty feet, during which the caves and old shore-lines were formed.
After this subsidence the islands were elevated to about their present
position. It only remains to sum up the facts that bear on the question
of the most recent movement in the Bahamas.

The fact that on the west coast of Andros, where the slope seaward
is so exceedingly slight, the soft calcareous mud grows gradually
harder and harder as we go inland, indicates to my mind that the island
has been recently rising ; for if it were subsiding or had recently sub-
sided, we may suppose that time enough would have elapsed since its
elevation to allow the calcareous deposit to harden into rock, and then,
as the subsidence took place, the surface at the edge of the water would
be hard rock, which would finally probably extend under the water as
the latter encroached on the land. The depth, close to the shore, of
the fine calcareous deposit also points to elevation, for in it I ran a pole
nine feet. How much deeper it was I had no means of determining,
owing to the length of the pole. Now, had subsidence been taking
place, should we find this depth of calcareous mud close to the shore ?
It is perhaps possible, and it might be claimed that the greater the sub-
sidence, the greater would be the depth of the mud ; but by the time
that nine feet of sediment had been laid down it is reasonable to sup-
pose that the mud on the shore would have hardened, and then should
we not find the water washing against a rocky shore ?

As we approach the west side of Andros from the interior, the pines

NOTES ON THE GEOLOGY OF THE BAHAMAS 43

grow smaller and smaller, and the forest is often prolonged into points
that run out in the swash and are composed of young and vigorous
trees. There are also in the swash small and slightly elevated pieces
of land barely above the level of the surrounding ground, and on these
are growing young pines. These facts show that the pines are advanc-
ing as fast as the conditions admit. Now, we may fairly suppose that
within very recent geological times the atmospheric conditions were
practically the same as now, and that the outward growth of the pines
is limited by the elevation of the land from the influence of the salt
water. Now, had the land once been as high as at present, it is fair to
suppose that the pines would have taken possession of the places they
now occupy; and if subsidence was in progress, they would now be
being driven back from their vantage ground. And we may also sup-
pose that before the change from elevation to subsidence had taken
place, time enough to allow the pines to grow old would have elapsed ;
and hence were the island now sinking, or rather had it recently been
sinking, we should expect to find on the western shore the pine forest
with an array of old and dying trees facing the sea, and it would not
be surprising if some dead pines were found standing in the water.
None of these appearances are to be found.

The mangroves, too, point to elevation, for I find recorded in my
note-books instances of mangroves, far above high-water mark, that
were apparently dying, but none were seen in situations that indicated
that the water was becoming too deep for them, as would probably be
the case if the land had been recently sinking.

These facts have perhaps been treated with too much detail; but
when I had been only to Nassau, and had seen the active erosion that
was taking place, I was inclined to think that the islands were sinking,
and was afterwards forced by the facts given above to alter my conclu-
sion. That most of the cays are being worn away and reduced in size
is evident. The vegetation is being driven backward, as indicated by
the old and gnarled appearance of the bushes and by the rhizomorphs
that remain where the plants once grew. But on the west side of Andros
the water is so shallow for a long distance seaward that there is practi-
cally no erosion of the land by the waves. A little bank of coral sand
and foraminifera raised slightly above the level of the swash is the only
evidence of the action of the water.

44 NOTES ON THE GEOLOGY OF THE BAHAMAS

Formation of the Cays north of New Providence

The position, size, and shape of these larger cays have already been
noted. Before attempting to explain the manner in which I believe
they were formed, I will briefly describe the appearance of the reef
at Nicol's Town, Andros, and also that at Rose Island, near Nassau.

At Nicol's Town the reef was about a quarter of a mile from the
shore, and the water varied from ten to eighteen feet in depth. The
reef was almost entirely composed of the great Madrepora palmarum,
with some small patches of M. cervicornus. Both of these corals were
growing luxuriantly, and their tips were close to the surface of the
receding waves. The reef at Rose Island, however, contained very
few specimens of M. palmarum, but many of M. cervicornus, and also
large heads of what are popularly known as ''brain corals," but which
of the genera so designated I am unable to state. The receding waves
left the top of the reef bare, and exposed the sea fans and alcyonoid
corals that were growing on the top. Toward the shore the reef sloped
gradually downward into a bank of sand, or, in other words, the sandy
bottom sloped upward to the top of the reef, which thus presented a
very different appearance from the one at Nicol's Town, where the
corals rose from the bottom. These facts show, I think, that the Rose
Island reef is an old one, while the Nicol's Town reef is in what we may
call its prime. A study of the chart of New Providence shows, running
along its northern shore, a number of reefs, some of which are marked
nearly dry at low water. Now, in order to convert these reefs into land
it is only necessary that they should be slightly elevated or that sand
should be deposited on top of them. And this, I believe, is the way
in which the cays known as Hog Island, Rose Island, Salt Cay, and
Quarantine Cay have been formed. In other words, I believe they
have been formed in the same manner as L. Agassiz explained the
formation of the cays of southeastern Florida. He argues, however,
that as the cays nearest the land, and hence those first formed, are no
higher than those farther seaward, it follows that during their formation
the land was stationary. Professor Heilprin,^ on the other hand, has
proved that the Florida peninsula is, or has lately been, rising. I
think I have shown that the Bahamas, or at least the islands of New

^ Trans. Wagner Free Inst. Sci., Vol. I.

NOTES ON THE GEOLOGY OF THE BAHAMAS 45

Providence and Andros, have very recently been rising; and yet,
although I have no measurements, I venture to state that Salt Cay, the
most seaward one, is as high if not higher than Hog Island, that lies
inside of it.

To return to Florida. It would seem at first as if the conclusions of
Professor Heilprin precluded the possibility of the cays being formed
as suggested by Agassiz ; for if we suppose a cay to be formed while the
land is rising, and then that another forms to seaward, and this process
to continue until three or four cays were formed one outside of the
other, it seems as if the first one would be higher above the level of the
sea than the last, and this Agassiz ^ states is not the case. When I first
read Agassiz' s memoir I was much impressed with the strength of his
arguments. But Heilprin's in regard to elevation are conclusive; and
although he did not visit the lower part of the peninsula, it is probable
that its movement is in the same direction as that portion examined.

It will, however, be noticed that Agassiz lays no stress on the eiffects
of the erosion that takes place so rapidly on all islands of coral forma-
tion.

I believe that the views of these two eminent observers can be
brought into harmony by considering the effects of erosion. Let us
suppose a cay has been formed of calcareous material thrown up by the
waves to a height of fifteen or twenty feet. Then, while the cay to
seaward was being formed, the erosion would take place on the one
already made, and whether its height above the sea increased, remained
constant, or decreased would depend upon the relative effects of the
elevation and erosion. It is not improbable that these two forces might
balance each other; and if this were so, it is easy to show diagrammati-
cally that any number of cays could form, one after the other, to sea-
ward, and yet the first be no higher than the last above the level of the
sea. And this argument also applies to the cays near Nassau. It is
probable that the ocean would throw up some of the cays higher than
the others, and also that a slight difference in height would not be
observed by the unaided eye.

1 "Report on Florida Reefs," Mem. Mus. Comp. Zool., Vol. VII, No. i, p. 36.

46 NOTES ON THE GEOLOGY OF THE BAHAMAS

The Formation of the Bahamas

As previously stated, the islands lie on the eastern and northern
edges of the Great and Little Bahama Banks. The only exceptions
axe Great Bahama Island, the Biminis, and some others that are
exceedingly small.

As I have visited only New Providence and Andros, I hesitate to
do more than suggest that the other islands of the banks have been
formed by the action of the waves and wind in the same manner as the
cays near Nassau.

A visit to the other islands of the group would be well repaid, for
some of them, as San Salvador, Rum Cay, and Inagua, have appar-
ently been formed in a different manner from those on the Great Bank.
And it is interesting to note that Dr. Bryant ^ describes Inagua as a
raised atoll. And Crooked Island and Acklin together resemble in a
remarkable manner the shape of the Keeling Atoll.^ Should these be
proved portions of an elevated atoll, it would be interesting, as they
would be only exceeded in size by the atolls of the Maldiva Archipelago,
which are the largest known.^ The islands of the Caicos Bank also
resemble the form of an old atoll, and the bank is still larger than that
on which Crooked and Acklin islands are situated.

In conclusion, it is interesting to note that Darwin and Professor
Dana, although they never visited the Bahamas, have, from a study of
the charts, come to diametrically opposite conclusions in regard to
the evidence they present of elevation or subsidence. Professor Dana
states* that ''the Bahamas show by their form and position that they
cover a submerged land of large area," and that " the long line of reefs
and the Florida cays trending away from the land of southern Florida
are evidence that this Florida region participated, though to a less
extent than the Bahamas. ... * Thus the size of the islands, as well
as the existence of coral banks and also the blankness of the ocean
beyond, all appear to bear evidence to a great subsidence."

Darwin, on the contrary, says:^ proofs of elevation within recent
Tertiary periods abound over nearly the whole area of the West

^ Proc. Bost. Soc. Nat. Hist., Vol. XI, 1866, p. 63.

2 Darwin, "Coral Reefs," 3d ed., PI. I, Fig. 10.

3 Dana, "Corals and Coral Islands," p. 190.
* "Corals and Coral Inlands," p. 368.

6 "Coral Reefs," 3d ed., p. 268.

NOTES ON THE GEOLOGY OF THE BAHAMAS 47

Indies, and hence it is easy to understand "the origin of the Great
Bahama Banks, which are bordered on their southern and western
edges ^ by singularly shaped islands formed of sand, shells, and coral
rock, some of them about loo feet in height, is easily explained by the
elevation of banks fringed on their windward side by reefs."

I think the facts I have given justify my conclusion in regard
to the recent elevation of Andros and New Providence. It is probable
that the elevation extended over the rest of the Bahamas, as caves
exist on the other islands. What the Bahamas are doing to-day, of
course, we cannot tell ; but until we have proof to the contrary, we
may assume that they are rising.

^ I have quoted this as given by Darwin. It is evidently a slip of the pen, as the
islands are on the eastern and northern edges of the banks.

BAHAMAN BIRDS

By John I. Northrop
(Abstract of address before the N. Y. Academy of Science)

After a few preliminary remarks upon the situation and size of the
Bahamas, the speaker stated that a paper giving the details of the
collection of birds would appear in the Auk for January. He stated
that all the species, seventy-four in number, were represented by the
specimens on the table, but that he would only call attention to the
most interesting.

The following birds were then exhibited and remarked upon:
Mimocichla plumbea, Mimus polyglottos, Mimus Gundlachi, Poliop-
tila ccerulea ccEsiogaster, Seiurus aurocapillus, Geothlypis rostrata (a
local form, slightly differing from the typical species found on New
Providence), Callichelidon cyaneoviridis, Doricha evelyncB, Sporadinus
ricordi, Chordeiles minor, Icturus northropi (a new species of a genus
not before reported from the Bahamas, and which has been described
and named by Dr. J. A. Allen), Coccyzus minor Maynardi, Saurothera
bahamensis, Phoenicopterus ruber (the habits and mode of capture
being described), Nycticorax nycticorax ncBvius (new to the Baha-
mas), Rallus coryi, Ardea bahamensis, and several other water birds.

The speaker also mentioned finding in the stomach of Antrostomus
carolinensis an entire humming-bird sufficiently undigested to identify
as Sporadinus ricordi.

He stated that the only mammals on Andros were bats, rats, and
mice. The bat was Macrotus waterhousei, and the rat Mus rattus.
The skin of the iguana, Cychera bcBolopha, was exhibited, and the
method of capture described. A few lizards were also shown, eight or
nine species having been collected. The speaker had collected a
number of species of snakes, the largest of which, a species of boa-
constrictor, was exhibited.

BAHAMAN BIRDS

He briefly mentioned the large collection of invertebrates, and
stated that, for the most part, it had not yet been worked up. One of
the Anemones, however, is probably new. In this connection the
speaker called attention to the interest connected with the geographi-
cal distribution of the Actiniaria, as Professor McMurrich considered
them to be related to Pacific forms.

In conclusion. Dr. Northrop mentioned the collection of plants,
comprising about seven thousand specimens, and representing about
six hundred to seven hundred species. These, he stated, were being
worked up by Mrs. Northrop, and twenty or thirty of the species
were in all probability new.

The plants, as well as the birds, are decidedly Antillean in char-
acter and affinities.

THE BIRDS OF ANDROS ISLAND, BAHAMAS^
By John I. Northrop

The island of Andros is the largest of the Bahama group, being
about ninety miles long, and forty or fifty miles across at the widest
part. The southern portion is separated from the northern by shallow
waters called "bights" ; but these are so filled with cays, as to make it
convenient to include all the islands under the general name of Andros.

Like all the others of the group, Andros is entirely of coral forma-
tion. The country is described by the natives as either "coppet,"
"pine-yard," or "swash." The first term is applied to the thicket
of angiospermous trees and shrubs that occupies the ridge along the
eastern coast. In most places this belt is very narrow, but near the
southern end it extends several miles into the interior. Back of the
coppet the land is comparatively level, and is covered by a forest of
the Bahama pine {Pinus hahamensis). As one approaches the west
coast, the pines become smaller and are mingled with palmettoes ;
finally both cease, and one sees spread before him thousands of acres
of level plain, supporting scarcely any vegetation except countless dwarf
mangroves. Here the ground is soft, and in wet weather almost en-
tirely under water; hence the peculiar appropriateness of the local
term "swash." Such is a brief description of the physical features of
Andros. As might naturally be supposed, the coppet proved the best
collecting ground for land birds, while the swash, and the lakes it
contained, were well stocked with many aquatic species.

Although Andros is the largest of the Bahama Islands, it seems
never to have been thoroughly explored by naturalists. The first
record of Bahama birds is given by Catesby^ who visited Andros.
In 1859 and again in 1866 Dr. Bryant made a collecting trip through
the Bahamas and published the results in the Proceedings of the
Boston Society of Natural History. His two papers are devoted to

^ The Auk, Vol. VIII, No. i, January, 1891, pp. 64-80.

' Natural History of Carolina,, Florida, and the Bahama Islands.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 51

ornithology and contain several references to Andros. Mr. C. J.
Maynard, in 1884, spent some time on Andros and has published a few
notes on its birds and described some new species.^ Mr. Cory also
made an ornithological trip through the Bahamas, during which he
visited the eastern coast of Andros. The results of his work are pub-
lished in a well-illustrated volume entitled the " Birds of the Bahama
Islands." Mr. Ingraham has also visited Andros, but I have been
unable to find any notes by him on its avifauna. These, I believe, are
the only naturalists who have ever honored Andros with a visit.

As this island is separated from both Florida and Cuba by nearly
the same distance (about one hundred and twenty miles), the writer
thought that a collection of its birds might prove interesting. It may,
however, be well to state that the object of the visit to Andros was not
to make a collection of birds, and that lack of time and assistance pre-
vented us from obtaining larger series. Before going to Andros, some
time was spent on the island of New Providence, the western end of
which is about twenty-five miles from the nearest point of Andros.
There a few birds were collected, and these will be referred to in the
list that follows. I was accompanied during the trip by Mrs. Northrop,
who made notes on the occurrence of the birds, and more particularly
on their songs and habits ; and many thanks are due her for aid in the
preparation of this paper. '

We reached the northern part of Andros on March 14, 1890, and
left the southern end on the 3d of July, during that time sailing almost
completely around the island ; exploring the majority of the creeks on
both the east and west coasts, and passing through the bights in the
central part. But the greater portion of the time was spent in the vari-
ous negro settlements that were scattered along the eastern coast, hence
we had ample opportunities for observation, and for obtaining what is
probably a quite complete collection of the birds of the island.

Dr. Allen has kindly annotated the list, and thanks are due to
him and to his assistant, Mr. Chapman, for aid in determining several
of the species.

The collection includes 286 specimens; and in it are represented
12 orders, 27 families, 56 genera, and 74 species. Of these one, that
Dr. Allen has done us the honor to name Icterus northropi, is new to
science; and Nycticorax nycticorax ncBvius is new to the Bahamas,

^ American Exchange and Mart and Household Journal, Vol. III.

52 THE BIRDS OF ANDROS ISLAND, BAHAMAS

while Geothlypis rostrata has hitherto been found only on New Provi-
dence.

Unless otherwise stated, all the birds mentioned in this paper were
actually obtained, and are now in the Museum of the School of Mines,
Columbia College. I have added to the usual common names the
local names of the birds.

I. Mimocichla plumbea Linn. Blue Thrasher. — This Thrush was com-
mon in the high coppet near Deep Creek, not far from the southeastern ex-
tremity of the island. Two specimens were also obtained near the northern end,
but none were seen in the pines or in the swash on the west side. The bird hops
about on the ground or on the lower branches of the trees, its black throat and red
legs rendering it quite conspicuous. The stomach of one examined contained
fruits. Those shot on June 28 were in condition to breed.

*2. Galeoscoptes carolinensis Linn. Catbird. — The Catbird was very
common at Nicol's Town near the northern end of Andros during March and April,
and the last specimen was seen at Mastic Point about May 23.

3. Mimus polyglottos Linn. — Locally known as Brown Thrasher and Mock-
ing-bird. My specimens were all collected near the northern end of Andros,
the first on March 25, the last on April 16. They were quite common about our
house at Nicol's Town, and their song could be heard at almost any hour of the
day. It was loud and varied, each syllable being usually repeated three times.
While on the western coast in June, we noticed one of these Mocking-birds, perched
on the top of a palmetto. He was singing, apparently with all his heart, and at-
tracted our attention by springing up into the air a few feet, then dropping to
his perch again. This he repeated three times, singing all the while ; we after-
wards saw another bird go through the same performance.

The stomachs of the specimens examined contained the remains of the
fruit of the gum elemi {Bur sera gummifera), white ants, and pieces of snail
shells.

[The three specimens referred to this species are indistinguishable from
M. polyglottos of the Carolinas or Florida. They are hence very unlike the small
form of Mimus from Inagua, recognized by Mr. Sharpe as M. elegans. — J.
A. A.]

4. Mimus gundlachi Cuban. — This was more common than the species
above described, as we found it wherever we landed. Its song is louder, clearer,
and more varied than that of M. polyglottos. Its food consists of fruits of various
kinds, but in the stomach of one specimen some small bones were found, probably
those of an Anolus. The ovaries of a specimen shot on May 15 were much en-
larged. The inhabitants of Andros do not distinguish between these two species,
calling both either Brown Thrushes or Mocking-birds.

* The star prefixed to a number indicates that the species was observed by Mr.
Scott at the Dry Tortugas. See note on page 87.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 53

[The series of seven specimens presents a wide range of variation in both size
and color. In the largest specimen the wing measures 4.73 inches, the tail 5.33;
n the smallest specimen the wing measures 4.23, the tail 4.60. In one specimen
the cheeks, sides of the throat, and the lower throat are thickly and heavily spotted,
and the streaks on the flanks are very broad. In another the cheeks and sides of
the throat are scantily barred, and the lower throat is without spots. The other
specimens are variously intermediate between these. The difference in size may
be in part sexual. — J. A. A.]

5. Polioptila caerulea caesiogaster Ridgw. Blue-gray Gnatcatcher. — This
species was very common in the low shrubs that grew in the pine-yard. It
was a most confiding little bird, and would sit within a few feet of you, twitching
its head from side to side and uttering its low, wheezy little song, apparently
very well pleased with its own efforts.

*6. Compsothlypis americana Linn. Parula Warbler. — This Warbler was
collected in the northern part of Andros on March 26 and April 19. It was
only seen in two localities, and was not common.

*7. Mniotilta varia Linn. Black-and-white Warbler. — First seen at NicoFs
Town on March 17. In a week or two they became quite numerous about
the house, but none were seen after the end of April. The three specimens pro-
cured were all males. Their stomachs contained the remains of beetles.

*8. Dendroica tigrina Gmel. Cape May Warbler. — This species was not
common. Specimens were collected on March 22, and on April 20, two on the
northeast coast, one near the west side.

9. Dendroica petechia Linn. Antillean Yellow Warbler. — But a single
specimen, a female, was obtained near Mangrove Cay on June 24.

*io. Dendroica caerulescens Gmel. Black-throated Blue Warbler. — One
specimen was brought to us in Nicol's Town, April 11; others were collected on
April 19 at Red Bay on the northwestern end of Andros. They were quite com-
mon about the house for a few days, but none were seen after the above date.

*ii. Dendroica striata Forst. BlackpoU Warbler. — The specimens were
all obtained at Conch Sound on May 19 and 20, although it was seen a little far-
ther south on May 23, and even a day or two later.

12. Dendroica vigorsii Aud. Pine Warbler. — This bird was one of the
most common species in the pine-yard. The five specimens collected all proved to
be females. The stomachs of those examined contained insects and small fruits.

13. Dendroica discolor Vieill. Prairie Warbler. — This was by far the most
common of the migratory Warblers. It was collected on the northern end of
Andros from March 22 to April 12, and a few were seen during the latter part of
April.

*i4. Dendroica palmarum Gmel. Palm Warbler. — A single specimen, a
female, was shot at Mastic Point, May 2; it was hopping about on the ground
under the small mangroves.

*i5. Seiurus aurocapillus Linn. Ovenbird. — This species was collected in
several localities on the northern end of the island. At Red Bay on the west side
it was quite common, and the natives knew the bird well under the name of the

54 THE BIRDS OF ANDROS ISLAND, BAHAMAS

** ground walker." All whom we questioned on the subject were certain that the
bird remained throughout the year, and some said that they had seen its nest.^
A specimen was collected by the writer on New Providence during January, and
the last time that we observed it was in the first week of May. The stomachs of
those examined contained the remains of insects.

[Four specimens taken on Andros Island, near the end of April, probably
represent a local resident form, differing slightly from the North American stock
in having the bill rather larger, the crown patch deeper orange, and the black
lines bordering it and the black streaks below slightly heavier. Should these dif-
ferences prove tolerably constant, they are too slight to render it desirable to des-
ignate the form in nomenclature. — J. A. A.]

i6. Geothlypis rostrata Bryant. Nassau Yellowthroat. — This species is new
to the island, all the specimens previously known being from the neighboring
island of New Providence. The first we saw was brought to us by a boy at Nicol's
Town, March 21. It was afterwards seen at Red Bays, Conch Sound, and later
at Mangrove Cay on June 25. It frequents the thick underbrush of the coppet,
and was always seen quite close to the ground. G. trichas was also collected, but
the difference between the two birds was very noticeable, G. rostrata being consid-
erably larger and much lighter in color. Its song was also louder and sweeter.

[The single specimen, male, is nearer G. rostrata than any other of the de-
scribed forms, differing from it in its much shorter and slenderer bill, with the
crown of a deeper and more bluish gray. Mr. Ridgway, who has examined the
specimen, says: "Intermediate between type of G. rostrata and G. tanneri ; bill
entirely like the latter; color above brighter olive-green than in either." While
doubtless representing a well-marked local race, it seems hardly worth while to
give it a name on the basis of a single specimen. — J. A. A.]

*i7, Geothlypis trichas Linn. Maryland Yellowthroat. — This species was
seen on both the east and the west side of Andros from April 12 to 19, but there
is no note of its appearance later. Of the five specimens collected, four were males.
It is interesting to note that Dr. Bryant states that of a flock of birds flying past his
vessel in the harbor of Grassy Creek, in the southern part of the island, on April 20,
1859, all were males.

*i8. Setophaga ruticilla Linn. American Redstart. — This Warbler was
first seen at Nicol's Town on April 10, and the last specimen was taken May 20;
one was seen a month later, however, on the west coast.

Since this paper was written an interesting article by Mr. W. E. D. Scott has
appeared in the Auk,^ on the "Birds observed at the Dry Tortugas, Florida,
during Parts of March and April." As I was collecting on Andros during the same
time, I thought it would be interesting to note the species common to both places
and have hence prefixed an asterisk to those mentioned in Mr. Scott's paper. A
reference to his paper will show that he secured many Warblers that I did not, and
a comparison of his dates with mine shows that with the exception of Mniotilta

^ Cory states that it is migratory. — "Birds of Bahamas," p. 71.
2 Vol. VII, p. 301.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 55

varia, Dendroica tigrina, D. caerulescens, and Geothlypis irichas, the birds were
observed later in Andros than at the Dry Tortugas. The commonest Warbler at
the latter place was D. palmarum, of which I only secured one specimen, and no
more were seen, while D. discolor, the commonest Warbler in Andros, was "not
uncommon" with Mr. Scott. I doubt, however, if any inference of importance
can be drawn from these facts.

In regard to the occurrence of the Warblers above mentioned it may be worth
while to state that they appeared in "waves." The most noticeable of these was
on April i8. We had been at Red Bays on the west side for a week, and had seen
very few birds about, but a bird wave must have arrived during the night of the
i8th, for the next day the grove about the house was full of birds, — the Black-
throated Blue, the Blue Yellow-back, the Redstart, Black-and-white Warbler, and
Vireo altiloquus barbatulus were seen here for the first time, and in considerable
numbers.

19. Coereba bahamensis Reich. Bahama Honey-creeper. "Banana Bird."
— Common throughout the island except in the swashes. The state of the genera-
tive organs of those taken near the end of March indicated that the birds were
nearly ready to breed.

*2o. Callichelidon cyaneoviridis Bryant. Bahama Swallow. — This beauti-
ful Swallow was first seen on New Providence and was afterwards found to be
abundant on Andros, particularly in the clearings. They generally flew quite
close to the ground. Dr. Bryant in the Proceedings of the Boston Society of
Natural History, Vol. VII, page in, says that of those collected by him up to
April 28, the genital organs exhibited no appearance of excitement. On April 15
I collected a male with the testes much enlarged. The people told me that the
Swallow breeds on the island, building under the rocky ledges, but we were not
fortunate enough to find a nest.

21. Vireo crassirostris Bryant. Large-billed Vireo. — Common at most places
on the island, and its cheery little song was almost constantly heard from the
bushes and low trees that the bird frequents. On May 24 the generative organs
of the male were enlarged.

[The four specimens seem distinctly referable to V. crassirostris. — J. A. A.]

*22. Vireo altiloquus barbatulus Cab. Black -whiskered Vireo. — The first
specimen of this bird was seen at Red Bays on the western side of the island, about
April 18. On our return to the east coast we found it common, and later noted it
as quite abundant at all our stopping places; and its song, so well described by
Dr. Bryant as "whip Tom Kelly, phew," was one of the most common notes of
the coppet. Dr. Bryant says that the seven specimens collected by him were all
males, and thinks that the females "had not arrived by the 13th of May." As I
collected the first female on May 24, having previously shot two males, this sur-
mise is probably correct. The genital organs were noted as being enlarged from
May 10 to June 5, when the last specimen was shot. Some of the birds were
mated and apparently ready to breed.

23. Spindalis zena Linn. — This bird, called by the natives "Tom James'
Bird" and "Robin Redbreast," was very common in the coppet and the pines. It

56 THE BIRDS OF ANDROS ISLAND, Bx\HAMAS

is one of the most conspicuous birds of the island, as well as one of the tamest,
being quite often caught by the children. They seemed to be particularly fond of
the fruit of the gum elemi {Bursera gummifera) and of the ripe figs. About May 20
the genital organs were enlarged.

*24. Piranga rubra Linn. Summer Tanager. — A single specimen, a male,
was taken at Red Bays, April 19. We showed it to some of the natives, but they
had never seen any like it. As its color would render it conspicuous, we may
infer that it was not a regular visitor, but, as stated by Cory in his "Birds of the
West Indies," p. 85, is only accidental in the Bahamas.

25. Loxigilla violacea Linn. Violet Grosbeak. — Known by the natives as
the "Spanish Paroquet," and quite abundant both in the coppet and the pines.
The males vary much in color, some being a brilliant black, while others are only
dusky; this difference is more apparent in living than in dead specimens. Both
sexes also vary in size. Some of these birds were brought to us alive by the negro
children, and I soon learned by experience that they not only were pugnacious but
could inflict quite a painful bite with their strong beaks. The generative organs
of a male collected May 30 were considerably enlarged.

26. Euetheia bicolor Linn. Grassquit. — Very common both in the pine-
yard and in the coppet, and the most abundant species of land bird on the island.
They were very tame and allowed one to approach within a few feet, and while we
were at Nicol's Town, several of them flew through the house at various times.
They seemed to vary a great deal, some of the males having the head and breast
black, while others were much lighter with only the throat and chin black, more
like the average female; in these the under mandibles were also much lighter.
They seemed to have mated even in April, as they were usually seen hopping
about in pairs.

27. Icterus northropi Allen. Northrop's Oriole. — (See PI. I.) Description
and notes of this bird were published in the Auk (Vol. VII, p. 343, October,
1890).

28. Agelaius phoeniceus bryanti Ridgw. Bahaman Red-winged Blackbird. —
This bird was found generally distributed over the island, and was the presiding
genius of its creeks and swashes. On the west coast especially his flute-like "okra-
lee" was the most familiar note. The birds were usually in small flocks, but there
always seemed to be a great preponderance of males. Beyond doubt, however,
the latter knew that their plainly dressed mates were near, much oftener than we,
for a number of times we saw one half raise his wings so as to expose his brilliant
epaulets, and sing with all his heart, fluttering his wings at every note. On two
occasions while visiting a large mangrove near Mastic Point, we remained until
after sunset and watched the Redwings fly out to roost there. The mangrove,
about two miles from the shore, was very large, and accommodated hundreds of
feathered guests nightly. There were Man-o'-war Birds, White-headed Pigeons,
and Louisiana Herons, but the most numerous as well as the noisiest lodgers were
the Red-winged Blackbirds. They began to arrive about five o'clock, flying in
from the mainland by twos and threes, or sometimes singly, until at last the por-
tion of the mangrove that they occupied was seemingly alive with them. There

THE BIRDS OF ANDROS ISLAND, BAHAMAS 57

must have been several hundred, males, females, and young, and they kept up a
constant chorus of hoarse chirps, screams, and gurgling "okralees" until quite a
while after the sun had set.

29. Pitangus bahamensis Bryant. Bahama Kingbird. "Fighter." — Not
uncommon in many parts of the island. I have nothing to add to Dr. Bryant's
account of its habits,^ except that in the stomach of one some fruits of the gum elemi
were found, showing that its diet was not entirely insectivorous.

30. Blacicus bahamensis Bryant. Bahaman Wood Pewee. — This small
species, of which Dr. Bryant says he only saw three specimens, all in the month of
March, we found common on the island, not only in the coppet, but also in the pine-
yard, and even occasionally in the swashes. At one of our stopping places two or
three were always to be seen flying about near the house. They were very fear-
less, and several times flew so close as to touch us. One seemed to be particularly
tame and would fly in at the door or window or about our heads with perfect un-
concern. He had several favorite perches about the house, on one or another of
which he was to be found at any time of day, cocking his little head from side to
side, in wait for unwary flies.

31. Myiarchis lucaysiensis Bryant. Rufous -tailed Flycatcher. — Not as abun-
dant as the preceding species, but quite often noted, especially in the northern part
of the island. Seven specimens were collected, all of which were males.

[The series of seven specimens agrees with one of Bryant's original specimens
in the Lawrence Collection at the American Museum of Natural History, and
differs from a small series of M. sagrce in being much larger, with the bill narrower,
and in lacking the strong yellow suffusion of the lower abdomen, crissum, and
axillaries, which in M. lucaysiensis are either nearly pure white or only faintly
tinged with pale yellow. It is apparently a strongly marked form. — J. A. A.]

*32. Tyrannus dominicensis Gwe^. Gray Kingbird. "Fighter." — Quite com-
mon on the island, but most abundant near the coast. On May 31 we found
a nest of this species in a small mangrove between tide marks. We had often
seen similar nests before, but could not be certain what bird had made them. The
nest was composed of small sticks rather compactly arranged. It was about nine
inches in diameter and four or five inches in depth. In the center was the cup,
about three inches in diameter and two or three deep, very neatly lined with a fiber
from some palm, probably from the spathe of the cocoanut palm. While exam-
ining the nest, the old birds flew from the shore and swooped over us, much
agitated. There were no eggs. On June 19 another nest was found in a man-
grove on the west side of the island. This time I shot both birds and secured
the nest and the one egg it contained. The latter was ovate in shape, 1.94 inches
long, and .70 wide; white with a few small brownish and lavender spots and
blotches, mostly near the larger end. It is stated that in Jamaica the nest of this
bird "is seldom found in any other tree than that of the palm kind." ^ In Florida,
however, it builds in the same situations as in the Bahamas. We found it a rather

1 Proc. Bost. See. Nat. Hist., Vol. VII, p. 108.

^ Baird, Brewer, and Ridgway, "North American Birds," Vol. II, p. 322.

58 THE BIRDS OF ANDROS ISLAND, BAHAMAS

quiet bird, although it has been described as noisy. Once we saw it attack a Buz-
zard, and persistently drive it away.

^^. Chordeiles minor Cab. Cuban Nighthawk. — Quite abundant; some-
times a dozen or more were seen flying about in the dusk. It is called "Pira-
midig" by the natives, on account of its cry, which to us, however, sounded more
like " Pir-ra-me-quick-quick." Those procured were shot in the daytime, on the
western side of Andros, on June 1 7. They were sitting motionless on the hot dry
plain, and did not fly until we were within a very few feet of them. Even when
disturbed they would alight again very shortly, and crouch close to the ground as
before.

*34. Antrostomus carolinensis Gmel. Chuck -will's-widow. — Three speci-
mens were taken, two of which were disturbed in the woods during the daytime, and
the other shot at dusk. Upon examining the stomach of the first one I shot, I
found, amid an indistinguishable mass of brownish matter, a small bone, about
half an inch long, that looked like the leg of a small bird. The next one examined
contained in its stomach the partially digested remains of an entire Humming-bird,
enough of which was preserved to identify it beyond doubt as Sporadinus ricordi.
"The remains of a small bird are said to have been found within the stomach of
one of this species." ^ One collected on May 15 contained remains of beetles and
winged ants in its stomach. The testes were much enlarged, being about half an
inch in length, but I am not sure that the bird breeds on the island.

35. Doricha evelynse Bourc. Bahama Woodstar. — Well distributed over
both New Providence and Andros, and as common in the pines as in the coppet.
They are both pugnacious and curious, for one day we watched one chase a Den-
droica discolor off a tree, and follow it some little distance, and while walking
through the woods one of these birds would often alight close by or hover over our
heads, as if examining us, and they flew through the house a number of times.
They seemed to alight on the branches quite as frequently as other birds, and we
often watched them perch and preen their feathers. Their little silvery trill was
a quite common note in the pines, and for a while we thought it the song of some
bird in the distance until we caught sight of the tiny songster almost above our
heads. They do not always confine their attention to flowers, for one day a Dori-
cha, after fluttering about the basket of flowers in my hand, made a dart at a good-
sized spider in a web close by, and to our surprise demolished it and was off again
in a moment. Of the numbers that we saw at New Providence, none were adult
males, and of those collected on Andros there was a large preponderance of females.

On one occasion we had the good fortune to see a male who was evidently dis-
playing his charms. The female was perched on a branch of a low shrub and
before her the male was performing. His wings were vibrating rapidly in the
usual manner, and thus supported in the air he swung rapidly to and fro, at the
same time rising and falling, a movement very difficult to describe, but almost
exactly like that of a ball suspended by an elastic thread that stretches and con-
tracts as the ball swings back and forth. This exceedingly graceful movement

^ Baird, Brewer, and Ridgway, "North American Birds," Vol. II, p. 413.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 59

was executed through a small arc for a few minutes, and then was suddenly-
changed. The male expanded his tail, showing the cinnamon of the webs, and
then threw himself rapidly and violently from side to side in an almost hori-
zontal line. During this latter part of the display a rustling sound was produced,
probably by the vibrations of the wings, and a few short sharp notes were
uttered. He then darted suddenly at the female who all the while had been sit-
ting apparently unconcerned, seemed almost to touch her with his bill, and then
flew rapidly away. Shortly afterward the female left, flying in another direction.
During all of the display the two birds were facing each other and not more than
six or eight inches apart, and the gorget and tail of the male were exposed to their
fullest advantage.

36. Sporadinus ricordi Gerv. Ricord's Humming-bird. — This was abundant
on Andros, where all my specimens were obtained, none being seen on New
Providence. Curiously enough, in this species, there seemed to be a great prepon-
derance of males, and out of the seven specimens collected, but one proved to be a
female. It may safely be said that most of those we saw were males, as the
females are quite different, being smaller and lacking the beautiful blue green
gorget. The ovaries of the bird collected May 16 were not enlarged. It is prob-
able that both Doricha evelynce and Sporadinus ricordi breed on the island; that
one of them certainly does was evidenced by a nest that I was shown. It was
lined with cotton, and was evidently the nest of a Humming-bird.

37. Crotophaga ani Linn. Ani. Rain Crow. — Frequently seen on both New
Providence and Andros, and always in small flocks of three or more. They
were not as tame as most of the land birds, being on the contrary quite shy, and
it was a difficult matter to get very close to them, as one of the flock was almost
certain to espy you, and to notify the others with his loud warning ^'wee-eep,'^ a
note in which the second syllable was much higher than the first, and more pro-
longed. On one occasion while walking through the woods during a rain, we no-
ticed what looked like a square black board suspended to a pine some distance
from the path. On investigation the black object was discovered to be seven
Crotophagas, sitting in a row on a small dead branch, and crowded as closely
together as possible. They were not asleep, however, for when I had approached
within thirty or forty feet, a warning note was heard, then another and another,
as they one by one took flight. '

38. Saurothera bahamensis Bryant. Great Bahama Cuckoo. — The "Rain
Crow," as the natives call this bird, was said to be abundant, but we got but one
mature specimen and two young ones. Dr. Bryant states that they were quite
common on New Providence in 1859. They fly about with a weak, wavy motion,
and the people told us that they could catch them after a short chase, as they soon
grew tired. Our specimens were obtained on the following dates, May 20, May 26,
June 13; all on the eastern side of Andros. The one taken on May 20 was a ma-
ture female, and the ovaries were enlarged, some of the ova being over one-fourth
of an inch in diameter. In the stomach were grasshoppers and a small lizard.
The stomachs of the others contained the remains of insects. Their bills were
soiled, probably from investigating the leaves of the epiphytic Tillandsias, as these

60 THE BIRDS OF ANDROS ISLAND, BAHAMAS

ants found in their stomachs seemed to be of the same species as those inhabiting
these plants. The description given by Cory ^ agrees with the two young speci-
mens, but the adult has a black band near the end of the tail-feathers, which are
tipped with dirty white, except the two middle ones. This agrees with the de-
scription given by Bryant.-

39. Coccyzus minor maynardi Ridgw. Maynard's Cuckoo. — This bird,
like the last species, was said to be common, but we succeeded in getting but four
specimens, and during our stay saw about as many more. Their note was quite
frequently heard, however, and usually from the mangroves or near by. The
stomachs contained the remains of small insects and grasshoppers. On June 28
the testes were much enlarged (one-half inch long).

*40. Ceryle alcyon Linn. Belted Kingfisher. — Not uncommon on either
New Providence or Andros, but no specimen was obtained. One was noted on
the west side on April 23, and another on the east side two days later, although Dr.
Bryant states that he saw none after April i. We thought that we saw one May 26,
but were not near enough in this instance to be certain.

41. Dryobates villosus maynardi Ridgw. Bahaman Hairy Woodpecker. —
All our specimens were taken on Andros, where the bird was abundant, especially
about the clearings.

*42. Sphyrapicus varius Linn. Yellow-bellied Sapsucker. — A single speci-
men, a female, was shot near Nassau about February 5.

43. Strix pratincola Bonap. American Barn Owl. — Although but few of
these birds were seen, I judge from what the people said that they are common.
One that I shot at Nicol's Town had its home under an overhanging ledge on the
precipitous side of a large hole about one hundred feet in diameter, known as the
''ocean hole." Here, in a deep recess, on March 25, we found two young ones not
yet able to fly; and near by were the remains of the common rat of the island {Mus
rattus). As is usual with this species, there was no nest, the birds resting on the
ground. Around them for some distance the ground was covered with the re-
jected food balls, composed of the bones and hair of the rodent above mentioned,
and as no other bones were noticed, it is probable that the Owl's principal article
of diet was rat.

44. Speotyto cunicularia dominicensis Cory. Burrowing Owl. — The speci-
men doubtfully referred to this subspecies was shot at Nassau in February.
Another, presumably of the same subspecies, was seen on the southern part of
Andros in June, flying about the low shrubs near the shore, but we were not for-
tunate enough to get it.

[A single specimen is referred to this form, which it much more resembles
than it does the Florida form, being much darker than the latter. — J. A. A.]

*45. Falco columbarius Linn. Pigeon Hawk. — A single specimen, a male,
was shot at Mastic Point on Andros on May 2. The bird was rare. A larger
Hawk also was seen, but no specimens were obtained.

* "Birds of West Indies," p. 159.

2 Pj-oc. Best. See. Nat. Hist., Vol. IX, p. 280.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 61

46. Cathartes aura Linn. Turkey Buzzard. " Crow." — Very common
on Andros, but more were seen on New Providence.

47. Columba leucocephala Linn. White -crowned Pigeon. — Very abundant
during the spring and summer. In the evening we often watched them flying
from Andi-os in the direction of New Providence. The people said that they were
going to Green Bay, a distance of fifty miles. A female shot on June 28 was
almost ready to lay.

48. Columbigallina passerina Ziww. — Called "Ground Dove," and "To-
bacco Dove" by the natives. Exceedingly common both on Andros and New
Providence. They generally fly about in small flocks of three to six, and are very
tame.

49. Zenaida zenaida Bonap. Zenaida Dove. — Not as abundant as the
White-crowned Pigeon, but not uncommon in the coppet. They were shy, and
were heard much oftener than seen.

50. Charadrius squatarola Linn. — A pair of Black-bellied Plovers was
shot on the shore near Red Bays, Andros, on April 14.

51. ^gialitis vocifera Linn. — The Killdeer was first seen near Fresh
Creek on the ist of June, and on the 7th two were shot. They both proved to be
males, with the testes enlarged to about .5 inch in diameter. The species was not
abundant, and but few were afterward seen.

52. -^gialitis wilsonia Ord. Wilson's Plover. — Abundant and very tame.
It was found on almost all the sandy beaches, where it would sometimes run along
just ahead of us for nearly half a mile, appearing quite to forget that it could get
out of our way by flying. This species is known to breed in the Bahamas.^

53. Haematopus palliatus Temw. American Oyster-catcher. "Sea-pie." —
One of these birds was obtained near Red Bay on April 15. It was feeding on
the extensive sand flats at low tide. Later, others were occasionally seen in similar
places. It is said to breed in the Bahamas.^

54. Himantopus mexicanus MUll. Black-necked Stilt. — A flock of these
birds was seen in one of the lakes on the west side of Andros on April 21, and a
pair were secured. We afterwards found them occasionally in the marshes. On
June 17 we saw quite a number of these birds in the swash near Wide Opening.
Our man said that this was the kind of place in which they bred, and later we
found a nest. It was simply a slight depression in the ground, and contained
four ovate eggs of an olive-green color, blotched with brown, measuring i^ to i^
inches by i^.

55. Symphemia semipalmata Gmel. Willet. — Commonly known as
" Tell-Bill-Willy." Very abundant in all the creeks and swashes. The ovaries
of those shot on May 31 were much enlarged.

*56. Actitis macularia Linn. Spotted Sandpiper. "Sandbird." — One
specimen, a female, was shot on the west side of the island, April 21, and was the
only one seen during our trip. As Cory obtained only three specimens in 1879,
it is doubtful if it is a resident.

^ Cory, "Birds of Bahamas," p. 14.

2 Bryant, Proc. Best. See. Nat. Hist., Vol. VII, p. 121.

62 THE BIRDS OF ANDROS ISLAND, BAHAMAS

57. Rallus coryi Maynard. Bahaman Rail. — A single specimen from
Conch Sound, Andros, April 15. It was shot and skinned by Mr. Alexander
Keith, a Scotch gentleman, to whom the writer takes pleasure in acknowledging
his indebtedness for this and many other favors, while on Andros. This bird was
known as the Marsh Hen, and is said by the people to be common, but we never
met with it again. On May 28 a woman brought us three eggs that she said were
those of a Marsh Hen. They are ovate to elongate -ovate in shape, and are nearly
cream color with chocolate spots and blotches irregularly distributed over the
entire surface but much more numerous at the larger end. Mingled with these
chocolate spots are others of a lavender gray. The eggs are of the following di-
mensions: 1.60 X 1.30; 1.70 X 1.22; 1.70 X 1.20 inches.

[A single specimen in fine (unworn) plumage is provisionally referred to Rallus
coryi. It is, however, about the size of ordinary R. crepitans, from which it dif-
fers in the gray edgings of the plumage, being much broader than in even extreme
examples of that form, resulting in a generally grayer effect. — J. A. A.]

58. Ardea herodias Linn. Great Blue Heron. "Arsnicker." — Not un-
common.

59. Ardea rufescens Gmel. Reddish Egret. — Abundant in the swashes.
The white phase of this bird was also common; many were seen, and three speci-
mens were collected.

60. Ardea tricolor ruficollis Gosse. Louisiana Heron. "Switching-neck."
— Two pairs were obtained at Mastic Point in May. They were breeding in a
large mangrove, about a mile and a half from the shore. These birds agree with
Ridgway's description ("Manual North American Birds," p. 131) except that the
upper part of the throat is not white, but rufous mixed with white. These also
agree with Cory's description of A. cyanirostris. But a specimen collected on the
western side of Andros answers to Cory's description of A. leucogastra, var. leuco-
prymna, and differs from the above specimens in the following points. The fore-
head and crown are much darker, being quite black; the throat has more rufous,
and the neck is darker. The nest of the Mastic Point birds was composed of
small mangrove sticks, laid so as to make a circular structure nine inches in
diameter and three or four deep. There was a slight depression in the top, in
which were placed a few sticks, parallel to each other, and on these were four eggs.
The latter vary in shape; some being ovate, while others are almost equally pointed
at both ends. They are nearly malachite green ^ in color, but with a slightly
bluer tinge. The measurements are as follows: 1.71-1.84 x i. 29-1.34 inches.

*6i. Ardea caerulea Linn. Little Blue Heron. — One specimen, shot at
Stafford Creek May 5. No more were seen, which seems curious, as Dr. Bryant
regarded this as the most common species of Heron; ^ and Cory states that it was
abundant during the winter, but no adults were taken by him.^ My remarks, how-
ever, only apply to Andros, and the bird might be common in other localities.

62. Ardea bahamensis Brewster. Bahama Green Heron. — Locally known

^ Ridgway, "Nomenclature of Colors."
2Proc. Bost. Soc. Nat. Hist., Vol. VII, p. 120.
8 "Birds of the Bahamas," p. 171.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 63

by the expressive name of "Poor Joe." We found it quite abundant in the
creeks and swashes, and at Fresh Creek collected what is very probably the young
of this species, hitherto undescribed. The top of the head is clove-brown with a
slightly greenish gloss, streaked with cinnamon-rufous. The rest of the head,
front of the neck, and the breast are white striped and mottled with sepia and
bistre. The back and the remainder of the neck are olive, having the feathers
edged with cinnamon-rufous. The tail is similar to the adult. Lower parts are
gray, the feathers being edged with white; and the scapulars and wings are clove-
brown, the coverts having an elliptical mark of wood-brown and being edged with
cinnamon-rufous; the rest of the wing-feathers having a deltoid mark of white at
the end. The bill is ochre yellow, darker above, and shading into black near the
end of the upper mandible. Legs olive.

*62. Nycticorax nycticorax naevius Bodd. "Golden." — The Black-
crowned Night Heron is new to the Bahamas, and is said by the people to be abun-
dant, although we did not see very many individuals. Two were secured at Conch
Sound, March 30.

*64. Nycticorax violaceus Linn. Yellow-crowned Night Heron. —
About as abundant as the last species. Both the above species of Nycticorax are
locally known as " Goldens," pronounced ''gaulings."

65. Phoenicopterus ruber Linn. — A few Flamingoes were seen on the western
coast of Andros in April; but later, in June, when the breeding season had com-
menced, we found them very abundant. A pair were shot on June 18, and they
were then ready to lay. We were told that one of their breeding places was near
Big Cabbage Creek, and a day was spent in an unsuccessful attempt to find their
nests. They were described to us, however, by reliable people, who also stated
that the birds sat upon their nests "like any other bird," and not with their legs
hanging down on either side. The birds are exceedingly shy, and in the open
swashes it is difficult to get within shot; for while stealing warily toward them,
you are very apt to hear a warning "gong, gong," and off flies the entire flock,
a streak of flame against the sky. The Flamingoes, when feeding, push their head
into the mud under water; and this fact is taken advantage of to secure them.
While the head is under water the negro walks rapidly forward, taking about ten
steps and then stopping. By that time the bird will probably lift its head and
look around. The negro stands motionless and screens his face with a branch of
a tree until the Flamingo, apparently satisfied that the new object is inanimate,
quietly resumes his feeding, and the negro advances as before. I was assured by
intelligent men that in this way they had sometimes captured the birds alive with
their hands. That the bird feeds in the manner described above is well shown by
its mouth, which is peculiarly adapted to sifting from the mud any moUusks or
crustaceans that might serve as food, and the gape of the bill when opened to its
fullest extent is only about half an inch. We were also told that a screen is some-
times built of palm leaves, and behind this the native can easily advance within
shot of the flocks. A large number of young birds are yearly destroyed by the
people for food. We ate the bodies of those we obtained and found the flavor most
delicious. We had for a long time been living practically upon flour and hominy,

64 THE BIRDS OF ANDROS ISLAND, BAHAMAS

and this may have made the Flamingo seem better than it really was; nevertheless
I think it would bear comparison with any of the much sought after game-birds.
The stomachs of the pair I obtained contained no fish, but many small shells and
much mud.

66. Dendrocygna arborea Linn. Tree Duck. — A flock of Ducks was
seen in a lake on the west side on April 21, but we could obtain no specimens. A
negro who was with us said that they were "Whistling Ducks," and, as Cory ^
states that this species was quite abundant on Andros, the negro was probably
right. I was told that earlier in the year the Ducks were very numerous on the
swashes. The ground about the lake mentioned was pitted with shallow holes
about two to four inches in diameter which our man said had been made by the
Ducks while feeding, when the place had been covered with water. I also saw a
smaller Duck near Wide Opening on June 17. In answer to my questions, I was
told that it was a "Summer Duck."

*67. Fregata aquila Linn. Man-o'-war Bird. — A number of these birds
had their roosting place in the large mangrove near Mastic Point/already spoken
of in connection with the Red-winged Blackbirds. We were told that formerly
they used to breed there, but that of late years they had gone farther from the
settlement.

*68. Pelecanus fuscus Linn. Brown Pelican. — A few seen, and one, a
young bird, obtained.

*69. Phalacrocorax dilophus floridanus Aud. Florida Cormorant. — Cory,
in his "Birds of the Bahamas," states that this is an abundant species, but in his
"West Indian Birds" says it is accidental in the Bahamas. We often saw it, and
on June 16, while sailing through the Northern Bight, found them breeding on a
small island known as Cormorant Cay. Here were old and young birds, some not
able to leave the nest, but the majority able to walk. Those still in the nest were
covered with a soft sooty down, and their gular sacks were pale yellowish white,
darker near the bill. The nests were about eighteen inches in diameter, and about
one foot in height, roughly constructed of sticks.

*7o. Anous stolidus Linn. Noddy. — A flock of these birds was seen near
Fresh Creek, on June 6. They were occasionally seen afterward.

*7i. Larus atricilla Linn. Laughing Gull. — Abundant. First noted early
in April.

*72. Sterna maxima Bodd. Royal Tern. — One specimen shot April 14
on Long Sound on the northern coast of Andros. A few more were seen, but the
bird was not common.

*73. Sterna fuliginosa Gmel. Sooty Tern. — Since our return a specimen of
this bird has been kindly sent to us by Mr. Alexander Keith of Andros. We saw
none of this species while on the island.

74. Sterna anaethetus Scop. Bridled Tern. — Abundant on a small bay near
Fresh Creek, where they breed. The people call this and the preceding species
"egg birds," and during the season collect and eat their eggs. They were not

" Birds of Bahamas," p. 183.

THE BIRDS OF ANDROS ISLAND, BAHAMAS 65

breeding when we were at Fresh Creek early in June, but they were almost ready
to breed.

*75. Sterna antillarum Less. Least Tern. — Abundant off Fresh Creek on
a small cay near the larger one occupied by the Bridled Tern; neither, however,
seemed to trespass on the ground of the other.

DESCRIPTION OF A NEW SPECIES OF ICTERUS
FROM ANDROS ISLAND, BAHAMAS^

By J. A. Allen

Through the kindness of Mr. John I. Northrop, of the School of
Mines, Columbia College, New York, I have the pleasure of making
known a new species of Icterus from Andros Island, one of the larger
islands of the Bahaman group. During four months spent recently
on Andros Island, Mr. and Mrs. Northrop devoted much attention to
birds, collecting about seventy species, among them several new to the
Bahamas, as well as the novelty about to be described, which adds not
only a new genus to the Bahaman fauna, but a new species to science.
The species is represented by nine specimens, three of which are adult
males, one immature male, one adult female, three immature females,
and another immature specimen of which the sex could not be deter-
mined. As shown by Mr. Northrop' s notes given below, he found the
species not uncommon. It is known to the residents of the island as
the ''Cocoanut Bird," and is said to be resident throughout the year.

Icterus northropi, sp. nov.

Adult Male. — Whole anterior half of the body, as far as the middle of the
breast below, and including the interscapulium above, together with the wings
(except the lesser and median coverts), and tail, deep black; rest of the body,
the thighs, lesser and median wing-coverts, edge of the wing, lower wing-coverts
and axillars, rich lemon-yellow; greater wing-coverts and primaries very narrowly
edged, and the outer tail-feathers very narrowly tipped with white. Bill and feet
black; lower mandible with the basal third bluish. In one specimen the longest
two lower tail-coverts are mixed yellow and black; in the other specimens they
are all wholly yellow.

Adult Female. — Similar to the male, except slightly smaller, and with the
black a little less lustrous.

Young. — Immature birds of probably the second year are olivaceous gray
above, brighter and more yellowish on the front of the head ; lesser wing-coverts,
lower back, rump, upper tail-coverts, and whole lower surface greenish yellow,

^ The Auk, Vol. VII, No. 4, October, 1890, pp. 344-346.
66

THE AUK, Vol. viii.

PLATE L

ICTERUS NORTHROPI allen.

1. ADULT MALE.

(% NAT. SIZE.)

2. YOUNG MALE.

(% NAT. SIZE.)

DESCRIPTION OF A NEW SPECIES OF ICTERUS 67

brightest on the rump and middle of the abdomen; median wing-coverts pale
sulphur-yellow; greater coverts brown, edged with whitish; chin, throat, and
cheeks much mixed with black, which here prevails over the yellow; there are
scattered black feathers over the breast and head, and in one specimen blackish
patches on the outer edge of the scapulars and sides of the breast; wings and tail
brown, the remiges edged with whitish, and the rectrices with olive.

Other specimens in a less advanced stage are similar, except that there are
fewer black feathers intermixed with the yellow, the chin and the front edge of the
cheeks alone being decidedly blackish, and the back is less olivaceous.

Measurements. — Length (from skins) .205 mm. (195-215 mm.); wing
94 mm. (90-99 mm.); tail 94 mm. (90-97 mm.); culmen 22 mm. (21-22 mm.);
tarsus 25 mm. (24-27 mm.).

Types. — No. 49,911, Am. Mus. Nat. Hist., d ad., Andros Island, Bahamas,
June, 1890; Mr. and Mrs. John I.i^Northrop. No. 49,912, Am. Mus. Nat. Hist.,
V ad., Andros Island, Bahamas, April 16, 1890.

The extent and distribution of the black and yellow in the adults
are the same as in P. wagleri except that the tail-coverts are yellow
instead of black; the tint of the yellow is nearly as in I. dominicensis,
from which it differs in having the whole lower parts yellow from the
middle of the breast posteriorly, instead of the yellow being confined
to the sides of the abdomen and crissum. The two species agree in
general size, but in /. northropi the bill is much stouter than in /.
dominicensis.

It is surprising that a bird so conspicuous, and apparently so com-
mon, as this should hitherto have escaped observation, Andros Island
having been several times previously visited by ornithologists.

Mr. Northrop has kindly presented the types of this species to the
American Museum of Natural History, and contributes the following
notes on its habits and distribution : —

" The above species of Icterus was first collected by us at Nicol's
Town, near the northern end of Andros, on April 8, 1890. We had
been there nearly a month when one morning a new note called us out
of the house, and we saw three or four of these birds flying about the
shrubs near by. They were so tame and unsuspicious that when one
was shot, the others kept their positions undisturbed until they met
a similar fate. Of the three we got then, one was a male, one a female,
and one we could not determine, but they were all in immature plu-
mage. Two weeks or so later, while on a trip to the west side, we saw a
number of these birds near Red Bays, and this time were fortunate

68 DESCRIPTION OF A NEW SPECIES OF ICTERUS

enough to get a male and female in full plumage. They were flying
about the palmettoes, and the flower stalk of an agave which was a mass
of golden blossoms and a great attraction to all the birds in the neigh-
borhood, as the flowers contained a large amount of nectar. While
near here one day, we heard a great commotion, and approaching
the scene, found two of these young birds fighting so violently that we
got almost near enough to take them up in our hands. The only
sound we heard them utter here was a rather plaintive call of two notes ;
but a month later, on May 22, we heard their song. It was a sort of
whistle of eight or nine notes, very sweet and pleasing, and almost
always given with the same intervals, and the same arrangement of
notes. The bird was observed in two other localities, the last time, June
18, near Wide Opening on the west side. It seems to inhg^bit the more
open portions of the island, near the coast, as we never saw any inland.

"The natives say that it builds its nest in the cocoanut trees, and
is always about them, hence its local name of ' Cocoanut Bird.' They
also told us that it remained throughout the year ; that the eggs were
pure white; and that the young birds differed from the old in plumage.
From the condition of the organs of generation, it seems that the bird
breeds during the month of June.

" The stomachs of those examined contained the remains of a grass-
hopper, beetles, and seeds ; hence it may be inferred that its food con-
sists of fruits and insects."

THE NORTHROP COLLECTION OF CRUSTACEA FROM

THE BAHAMAS '

By W. M. Rankin

(Pis. XXIX-XXX)

The Crustacea collected by Dr. and Mrs. Northrop in the Ba-
hama Islands in 1890 were sent to me by Professor Osborn, with
the request that I prepare a report on them. The following list is the
result. Such a list is of necessity largely a mere catalogue of names,
but it is hoped that it may be of service in the preparation of a more
extensive fauna of the Bahamas when such a work shall be undertaken.
It has been with the idea of giving a little wider interest to the list that
with each species the range of distribution has been given, and also
the West Indian islands noted where the species has been found, al-
though this latter record is no doubt incomplete. I hope at least these
notes of distribution may serve as a suggestion for the fuller record of
the distribution of these species among the West Indies. The synon-
ymy I have made brief, merely citing the original author and usually
a reference to the work where a complete synonymy may be found.

The letters (a), (b), etc., in many species indicate the various series
of specimens in the collection as they were arranged originally or, in
some cases, sorted out by me after their receipt. To these series I have
fortunately been able to add some notes made by Dr. Northrop when
the collections were made, and recently sent me by Mrs. Northrop.

Among the sixty-seven species collected I have determined four as
new species, and one I have ranked as a new variety. There is also
published for the first time a figure of Stenopus IcBvis. For the careful
drawings of the figures I am indebted to Mr. R. Weber. I wish to
express my obligations to Miss Rathbun, of the National Museum, for
assistance in identifying a few species; and also to Dr. Ortmann, of
Princeton, who has kindly assisted me in many ways and to whom this
report owes much of any value it may possess.

^ Ann. N. Y. Acad. Sci. XI, No. 12, August, 1898, pp. 225-258.

70 CRUSTACEA FROM THE BAHAMAS

DECAPODA

BRACHYURA-CATOMETOPA

Family Ocypodidae Ortmann

I. Ocypode arenaria (Catesby)

Cancer arenarius Catesby, History of the Carolinas, II, p. 35, 177 1.

Kingsley, Proc. Acad. Nat. Sci., Phil., 1880, p. 184.

Ortmann, Z06.I Jahrb., VII, p. 765, 1894.

(a) 5 ^ , 2 $ . Near Nassau, N.P., Jan. 24, 1890.

Range : South shore Long Island to Rio Janeiro.

Collected at Cuba, Jamaica, St. Thomas, New Providence.

2. Uca platydactyla (Milne-Edwards)

Gelasimus platydactylus Milne-Edwards, Hist, des Crustaces, II,
P- 51. 1837.

G. heterocheles Kingsley, I.e., 1880, p. 137.

(a) 4 $ . Under sides of stones, Dix Point, near Nassau, N.P.,
Feb. 4, 1890.

(^')8^,7?.

Range : East and west coasts Central America, West Indies.

Collected at Jamaica.

3. Uca vocator (Herbst)

Cancer vocator Herbst. Naturder Krabben u. Krebse, III, pt. IV,
1804.

Gelasimus vocator Martens; Kingsley, I.e., 1880, p. 147.

(a) I Z. Bahama Islands.

Range: East coast of America, west coast of Mexico, Panama,
West Indies.

Collected at Bahamas, Cuba, Hayti, Jamaica.

4. Uca stenodactyla (M. Edwards et Lucas)

Gelasimus stenodactylus , M. Edwards et Lucas in D'Orbigny's
Voyage, 1843.

Kingsley: I.e., 1880, p. 154. Ortmann: I.e., p. 760, 1894.

(a) I ^ . Common in mud on west side of Andros Island, near Red
Cays, April 17, 1890.

CRUSTACEA FROM THE BAHAMAS

Range : West Indies, Central America, east and west coast. -
Collected at Cuba.

5. Uca leptodactyla (Guerin Ms.)

Gelasimus leptodactylus Guerin Ms. (types in Phila. Acad.).

Gelasimus stenodactylus Kingsley, Proc. Acad. Nat. Sci., Phila.,
p. 155 (part), 1880.

(a) 10 ^ , 5 9 . Holes in sand between tides about 5-6 in. deep,
very shy, near Ft. Montagu, Nassau, N.P., Jan. 28, '90.

Some of these specimens were sent to the United States National
Museum, where they were identified by Miss Rathbun, and to whom
I am indebted for the following note of description :

" Uca leptodactyla belongs to the division of the genus in which the
front between the eyes is broad and the body is short, broad and sub-
cylindrical. It is most nearly related to U. stenodactyla ; the chief dif-
ferences are as follows : in U. stenodactyla the body is much higher
than in leptodactyla, being usually higher than long. The anterior
margin of the carapace from the base of the eyestalk to the antero-
lateral angle is much more oblique in leptodactylo, and the lateral mar-
gins are much more convergent posteriorly. The carapace of lepto-
dactyla is, therefore, more pentagonal than that of stenodactyla. In
stenodactyla the lateral margin is much dilated behind the antero-
lateral tooth, which is not the case in leptodactyla. The inner surface
of the hands differs as follows : the short ridge on the palm at the
base of the dactylus is perpendicular to the base of the propodos in
leptodactyla; while it is oblique in stenodactyla. In both species the
tubercular ridge running obliquely upward from the lower margin
makes an angular turn at the middle of the inner surfaces, and is
continued until near the upper margin. In leptodactyla this continua-
tion runs parallel to the line of tubercles at the base of the dactylus ;
in stenodactyla the continuation is directed obliquely towards the line
at the base of the dactylus."

Family Gecarcinidae Dana

6. Gecarcinus ruricola (Linnaeus)

Cancer ruricola Linnaeus, Sys. Nat. Ed. 10, 1, p. 626, 1758.
Gecarcinus ruricola Leach. Edin. Encyc, VII, 430, 1814. Ort-
mann. I.e., p. 740, 1894.

72 CRUSTACEA FROM THE BAHAMAS

(a) I $ , Bahama Islands. (Dry.)

Q)) 1 $ , Nicol's Town, Andros Island, March 9, 1890. (Dry.)

Range : West Indies, Mexico.

Collected at Cuba, Jamaica, Hayti, Martinique.

7. Cardisoma guanhumi (Latreille)

Latreille, Ency. Meth., Hist. Nat. Insectes, X., 685, 1825.
Ortmann, I.e., p. 735, 1894.

(a) I ^ , I $ , $ I, juv. Move sluggishly, make holes in the ground
by side of road under trees, Nassau, N.P., Jan. 25, 1890.

Range : East and west coasts of Central America, West Africa. •
Collected at Cuba, Jamaica, Hayti, St. Thomas, Barbadoes.

Family Grapsidae (Dana)

8. Leiolophus planissimus (Herbst)

Cancer planissimus Herbst, I.e., p. 3, PI. LIX, 1804.
Miers, Ann. Mag. Nat. Hist., Ser. 5, 1, 1878, p. 153.
(a) 3 ^ , I $ . On shore, just south of Ft. Montagu, Nassau, N.P.,
Jan. 22, 1890.

{h) 2 $ juv. Ocean side of Salt Cay, N.P., Jan. 31, 1890.
Range : '' Cosmopolitan, except the colder seas.'* Ortmann.
Collected at Jamaica.

9. Plagusia depresse (Fabricius)

Cancer depressus Fabricius, Entom. Sys. Suppl., p. 406, 1775.

Miers, Challenger, Brachyura, p. 272.

{a) 2$ . Salt Cay, New Providence. (Dry.)

Range : Charleston to Brazil, Mediterranean to St. Helena.

Collected at Cuba, Jamaica.

10. Sesarma cinerea (Say)

Sesarma ricordi Milne-Edwards, Annal Sci. Nat. (3) Zool. t. 20,
p. 183, 1853.

Ortmann, Carcinologische Studien, Zool. Jahrb., Bd. X, p. 329,
1897.

{a) I $ with ova. Under side of stones, Dix Pt., near Nassau,
N.P., Feb. 4, 1890.

CRUSTACEA FROM THE BAHAMAS 73

Range : West Indies.

Collected at St. Domingo, Hayti, Jamaica, St. Thomas, Virgin
Islands.

II. Pachygrapsus transversus (Gibbes)]

Gibbes, Proc. Am. Ass. Adv. Sci. Ill, p. i8i, 1850.

Kingsley, I.e., 1880, p. 198.

(a) 4 ^ , juv., 3 ? with ova. Nassau, N.P., under stones Jan.,
1890.

Range: ''Warm and temperate waters of both hemispheres."
Ortmann.

Collected at Cuba, Jamaica, Virgin Islands, Barbadoes.

12. Grapsus grapsus (Linnaeus)

Cancer grapsus Linnaeus, Sys. Nat. ed. X, I, p. 630, 1758.
Smith, Trans. Conn. Ac, IV, 1880, p. 256. Ortmann, I.e., p.
703, 1894. Kingsley, 1880, p. 192; 1879, p. 401.
(a) I ^ , 2 9 . Near Nassau, N.P., Jan., 1890.
Range : Warm waters of both hemispheres.
Collected at Cuba, Jamaica, Hayti.

13. Goniopsis cruentatus (Latreille)

Grapsus cruentatus Latreille, Hist. Nat. des Crust., VT, p. 70, 1803.

Kingsley, I.e., 1880, p. 190. Ortmann, I.e., p. 701, 1894.

{a) It. (Dry.)

(&) 2 !? . On shore near Nassau, N.P., Jan. 23, 1890.

Range : American and African coasts of the Atlantic Ocean.

Collected at Cuba, Jamaica, Hayti.

BRACHYURA-CYCLOMETOPA

Family Oziidae (Ortmann)
14. Eriphia gonagra (Fabricius)

Cancer gonagra Fabricius, Sp. Ins., p. 505, 1781.
Ortmann, I.e., p. 480, 1894.

(a) I $ . In pools on shore, Nassau, N.P., Jan. 21, 1890.
{h)i ? . Dix Point, near Nassau, N.P., Feb. 4, 1890.
(c) I ^. Salt Cay. Ocean side, near New Providence, Jan. 31,
1890.

74 CRUSTACEA FROM THE BAHAMAS

Range : Atlantic coast from Carolina to Rio Janeiro.
Collected at Bahamas, Cuba, Jamaica, Hayti, Barbadoes.

15. Domoecia hispida Eydoux et Sbuleyet

Eydoux et Souleyet, Voy. Bonite, I, Crust., p. 235, 1842. Ortmann,
I.e., p. 478, 1894.

(a) I $ , juv.

Range : West Indies, Florida, Cape Verde Islands, Senegal, Pacific
islands.

Collected at Cuba, Jamaica, St. Thomas, Guadaloupe.

16. Panopeus herbstii Milne-Edwards

Milne-Edwards, Hist. Nat. Crustaces, I, p. 403, 1834.

Benedict & Rathbun, Proc. U. S. Nat. Mus., XIV, p. 358, 1891.

(a)i$. Nassau, N.P., Jan., 1890.

Range : ^ Rhode Island to Brazil.

Collected at Bahamas, Jamaica, St. Thomas, Curasao, Trinidad.

17. Panopeus occidentalis Saussure

Saussure, Rev. et Mag. de Zool. (2), IX, p. 502, 1857.
Benedict & Rathbun, I.e., p. 360.

(a) I 9 . Near Nassau, N.P., Feb., 1890.

(b) 1 $ , On shore near Nassau, N.P., Jan. 22, 1890.
Range : Atlantic from South Carolina to Brazil.

Collected at Jamaica, Old Providence, Guadaloupe, Curasao,
Trinidad.

18. Panopeus americanus Saussure

Saussure, Rev. et Mag. de Zool. (2), IX, p. 502, 1857.
Benedict & Rathbun, I.e., p. 380.

(a) I ^,4 <?. Near New Providence, Bahamas, Jan.-Feb.,
1890.

{h) I $ . On shore, near Nassau, N.P., Jan. 22, 1890.
{c)i$ . Nassau, N.P., Feb. 24, 1890, Dix Point.
Range : West Indies to Brazil.
Collected at Jamaica, St. Thomas.

CRUSTACEA FROM THE BAHAMAS 75

Family Xanthidae Ortmann
19. Chlorodius floridianus Gibbs

Gibbs, I.e., p. 175, 1850.

(a) I 9 . Collected in pools and under stones, New Providence,
and neighboring cays.

(^) I $. Dix Point, Nassau, N.P., Feb. 24, 1890.

(d) 1 <? , 3 9 . On shore near Nassau, N.P., Jan. 22, 1890.
Range : Florida to Brazil.

Collected at Jamaica, St. Thomas, Barbadoes.

20. Lophactaea lobata (Milne-Edwards)

Cancer lobatus Milne-Edwsivds, Hist. Nat. Crustaces, I, p. 375, 1834.
LophactcBa lobata A. Milne-Edwards, Nouv. Arch. Mus. Hist. Nat.,
I, p. 249, PI. XVI, fig. 3, 1865.

(a) I ?. Quarantine Station, Jan. 25, 1890.
Range : West Indies, Gulf of Mexico, Bermuda.
Collected at Jamaica and the Antilles.

21. Heteractaea ceratopa (Stimpson)

Pilumnus ceratopus Stimpson, Ann. Lye. Nat. Hist. N.Y., VII,
p. 215, 1862.

Heteractcea ceratopus Kingsley, I.e., 1879, p. 396.
(a) I 2 . Dix Point, Nassau, N.P., Feb. 24, 1890.
(6) I $ . Quarantine station, N.P., Jan. 25, 1890.
Range : Florida and West Indies.
Collected at Guadaloupe.

22. Actaea acantha (Milne-Edwards)

Cancer acanthus Milne-Edwards, Hist. Nat. Cr., I, p. 390, 1834.
ActcBa acantha A. Milne-Edwards, Nouv. Arch., I, p. 278, PL XVI,
1865.

(a) I $ . Quarantine station near Nassau, N.P., Feb. 10, 1890.
Range : Florida Keys, West Indies.
Collected at Jamaica, Guadaloupe.

76 CRUSTACEA FROM THE BAHAMAS

PORTUNINEA

Family Portunidae (Ortmann)
23. Callinectes larvatus Ordway

Ordway, Boston Jour. Nat. Hist., VII, p. 573, 1863.

Rathbim, The genus Callinectes, Proc. U. S. Nat. Mus., XVIII ,j
p. 358, 1896.

(a) I ^ , I $ , spur. juv. On shore, just south of Ft. Montagu,
Nassau, N.P., Jan. 22, 1890.

Range : Florida to Brazil, West Indies, Cape Verde Islands, Africa.

Collected at Bahamas, San Domingo, Jamaica, St. Thomas.

24. Callinectes tumidus (Ordway)

Ordway, I.e., p. 574, 1863.

Rathbun, I.e., p. 359, 1896.

(a) I £^. Nassau, N.P., Jan. 21, 1890, common in shoal water.

Range : Florida to Brazil, West Indies.

Collected at Jamaica, Hayti, Old Providence.

25. Acheloiis depressifrons (Stimpson)

Amphitrite depressifrons Stimpson, Ann. Lye. Nat. Hist. N.Y.,
VII, p. 58, 1862.

Acheloiis depressifrons Stimpson, ibid., p. 223.

{a) 1 $ . Quarantine station, N.P., Jan. 25, 1890.

Range : South Carolina to Florida, Bermuda, West Indies. Be-
sides this specimen from New Providence, the Princeton Museum pos-
sesses one from the Virgin Islands; the only two localities reported
from the West Indies.

26. Acheloiis ordwayi (Stimpson)

Stimpson, Notes on N. Am. Crustacea, Ann. Lye. Nat. Hist.,
N.Y., p. 224, 1862.

Smith, Trans. Conn. Acad., II, p. 9.

(a) I $ . Quarantine station, N.P., Jan. 25, 1890.

{h) 1 $ , with ova. Dredged near Nassau, N.P., Jan. 22, 1890.

Range : Florida and West Indies.

Collected at St. Thomas.

CRUSTACEA FROM THE BAHAMAS 77

27. Acheloiis tmnidulus Stimpson

Stimpson, Bull. Mus. Comp. ZooL, II, p. 149, 1870.

{a) I $. Dredged near Nassau, N. P., Jan. 22, 1890.

Stimpson describes two specimens from the coast of Florida.
The species is probably only the young of A. ordwayi, as it only
differs from the latter (as noted by Stimpson) in the less prominent
frontal spines.

MAIOIDEA

Family Periceridae Miers
28. Macrocoeloma eutheca (Stimpson)

Pericera eutheca Stimpson, Bull. Mus. Comp. ZooL, II, p. 112, 1870.
Rathbun in Proc. U. S. Nat. Mus., Vol. XV, No. 901, p. 251, 1892.
(a) I $ . Dredged near Nassau, N.P., Jan. 22, 1890.
Range: Florida, West Indies.
Collected at Cuba.

29. Microphys bicomutus (Latreille)

Pisa hicornuta Latreille, Encyc. Meth., Hist. Nat. Insectes, X,
p. 141, 1825.

Microphys bicomutus A. Milne-Edwards, Nouv. Arch. Mus. Hist.
Nat., VIII, p. 247, 1872.

Rathbun, I.e. (No. 901), p. 253.

(a) 2 ^ , 5 $ . Common under rocks between tides and in pools,
New Providence, Jan., 1890.

{h) 1 t. Quarantine station, N.P., Jan. 25, 1890.

(d) 1 $ . On shore near Nassau, just south of Ft. Montagu, Jan.
22, 1890.

{e) I $ , juv. "Sea gardens," near Nassau, N.P., Feb., 1890

(f) 1 $, juv. Ocean side of Salt Cay, Feb. 6, 1890.

(^) I $, juv. Nassau, N.P., Feb. 15, 1890.

(h) 1 $ , juv. Salt Cay, N.P., ocean side, Jan. 31, 1890.

Range : Florida, West Indies to Brazil, Bermuda.

Collected at numerous islands of the West Indies.

78 CRUSTACEA FROM THE BAHAMAS

30. Othonia aculeata (Gibbes)

Hyas aculeata Gibbes, I.e., p. 171, 1850.
Rathbun, I.e., p. 255, 1892.
(a) I ^. On shore just south of Ft. Montagu, Nassau, N.P.,
Jan. 22, 1890.

Range : Florida and West Indies.

Collected at Cuba, Bahamas, Jamaica, St. Thomas, Guadaloupe.

31. Othonia Iherminieri Schramm

Schramm, Crust, de la Guadaloupe, 20, 1867.
{a) 1 $ , 2 $ . On shore near Nassau, Jan. 22, 1890.
The three specimens in the collection are broken and imperfect.
I place them doubtfully in this species.

Range: Atlantic coast; South Carolina to Brazil.
Collected at Cuba, Jamaica, St. Thomas, Guadaloupe.

32. Mithrax pilosus Rathbun

Rathbun, I.e., p. 262, PI. XXXIX (No. 901), 1892.

(a) 1 $. Near New Providence, Jan., 1890.

(b) 2 $, fragmentary. Salt Cay, ocean side, New Providence,
Jan. 31, 1890.

Miss Rathbun' s four specimens were collected in Abaco, Bahamas.

33. Mithrax cinctimanus (Stimpson)
Mithraculus cinctimanus Stimpson, Ann. Lye. Nat. Hist. N.Y.,
VII, p. 186, 1862.

Rathbun, I.e., p. 268 (No. 901), 1892.

(a) 1 $ . Dix Point, Nassau, N.P., Feb. 24, 1890.

(b) 1 $. Quarantine station, N.P., Jan., 25, 1890.

(d) I Q , juv. Nassau, N.P., Jan., 1890.

Range : Florida coast. West Indies, Gulf of Mexico.

Collected at Andros Island, Jamaica, St. Thomas, Guadaloupe.

34. Mithrax forceps (A. Milne-Edwards)
Mithraculus forceps A. Milne-Edwards, Miss. Sci. au Mexique,
pt. 5, 1, p. 109, 1875.

CRUSTACEA FROM THE BAHAMAS 79

Rathbun, I.e., p. 267 (No. 901), 1892.

(a) 6 $ , mostly young. Ocean side of Salt Cay, Feb. 6, 1890.

(b) 1 $ , fragmentary. Nassau, N.P.

(d) 1 $,2 $, juv.

Range : From North Carolina to Brazil and Guiana.
Collected at Nassau, Bahamas, Old Providence, St. Thomas,
Curasao.

35. Mithrax sculptus (Lamarck)

Maia sculpta Lamarck, Hist. Anim. sans Vert., V, p. 242, 1818.
Rathbun, I.e. (No. 901), p. 271, 1892.
{a) 1 $ ,1 $ . Quarantine station, N.P., Jan. 25, 1890.
Range : Florida, West Indies to Venezuela, Surinam.
Collected at numerous localities in the West Indies.

36. Mithrax coronatus (Herbst)

Cancer coronatus Herbst, Natur. der Krabben u. Krebse, I, p.
184, PL XI, fig. 6^, 1785.

Rathbun, I.e. (No. 901), p. 272, 1892.

{a) 1 $. Salt Cay, ocean side, near New Providence, Jan. 3, 1890.

(b) 1 $ , juv. Ocean side of Salt Cay, Feb. 6, 1890.

Range: Florida, West Indies, Central America, Brazil.

Collected at Abaco, Bahamas, Jamaica, Cuba, St. Thomas, Guad-
aloupe.

Family Inachidae Miers

37. Acanthonyx petiverii Milne-Edwards

Milne-Edwards, Hist. Nat. Crust., I, p. 343, 1834.
(a) 1 $, broken. Under rocks, between tides and in pools. Nassau,
N.P., Jan., 1890.

Range : West Indies to Brazil and California to Chili, Galapagos.
Collected at Cuba, Jamaica, St. Thomas, Guadaloupe, Martinique.

DROMIIDEA

Family Dromiidae Dana

38. Dromidia antillensis Stimpson

Stimpson, Notes on N. Am. Crust., Ann. Lye. Nat. Hist. N.Y.,
VII, p. 71, 1862.

80 CRUSTACEA FROM THE BAHAMAS

(a) I $ . Nassau, N.P., Feb. 15, 1890.
Range : Florida, West Indies, Brazil.
Collected at Antilles, Jamaica, St. Thomas.

HIPPIDEA

Family Hippidae Stimpson
39. Remipes cubensis Saussure

Saussure, Rev. et Mag. de Zool. (2), IX, p. 503, 1857.

Ortmann, Die geog. Verbreit. der Decap. gruppe der Hippidea,
Zool. Jahrb. IX, p. 219, 1896.

Remipes scutellatus (Fabricius) , Henderson, Challenger Anomura,
Z^, 1888.

{a) 19 5 with ova. Beach at Nicol's Town, Andros Island, April
4, 1890.

(&) 2 ^ , 9 <? . Quarantine station near New Providence, Jan.
25, 1890.

Range: ''American and African shores of Atlantic.'^ Ortmann
(1. c. supra) .

Collected at Cuba, Jamaica, St. Christophers, Barbadoes.

GALATHEIDEA

Family Porcellanidae Henderson
40. Porcellana sayana Leach

Pisidia sayana Leach, Diet. d. Sci. Nat., XVIII, p. 54, 1820.
Porcellana ocellata Gibbes, I.e., p. 190, 1850.
Henderson, Challenger, Anomura, p. 109, 1888.
{a) 1 $ . Came out of a shell inhabited by a large hermit crab.
Nassau, N.P., Jan. 26, 1890.

Range : West Indies and Southern shores of United States.
Collected at Antilles, Jamaica, St. Thomas.

41. Pachycheles panamensis Faxon

Faxon, Mem. Mus. Comp. Zool, XVIII, p. 75, Tab. 15, 1895.

Ortmann, Zool. Jahrb., X, 1897, p. 293.

(a) I ^ , 2 <? . Ocean side of Salt Cay, Feb. 6, 1890.

CRUSTACEA FROM THE BAHAMAS

Size of ^ 5 J mm. long, 5 mm. broad; of $ 5 mm. long, 6 mm.
broad. These specimens have been kindly examined for me by Dr.
Ortmann, who finds them identical with Faxon's type from Panama,
and also very close to the Cape Verde P. harhatus A. Milne-Edwards.
This is the first recorded specimen of P. panamensis from the West
Indies.

42. Petrolisthes armatus (Gibbes)

Porcellana armata Gibbes, I.e., p. 190, 1850.

Petrolisthes armatus Stimpson, Ann. Lye. Nat. Hist. N.Y., VII,
p. 73, 1862.

Ortmann, Zool. Jahrb., X., 1897, p. 280.

(a) I ^,1 <?. Ocean side of Salt Cay, Feb. 6, 1890.

Ortmann (I.e. supra) gives full synonymy of this species and makes
its distribution circumtropical ; West Indies to Brazil, Gibraltar, Cali-
fornia to Panama, Indo-Pacific.

Collected at Cuba, Jamaica, St. Thomas, Barbadoes.

43. Petrolisthes tridentatus Stimpson

Stimpson, Ann. Lye. Nat. Hist. N.Y., VII, p. 75, PI. I, 1862.
(a) I $ . Along shore, near Nassau, N.P., Feb. 20, 1890.
(6) 2 ^ , 5 $ . Salt Cay, N.P., ocean side, Jan. 31, 1890.
{c) 2 $ ,1 $. Under sponges, Nassau, N.P., Jan., 1890.
Range: West Indies.
Collected at St. Thomas, Barbadoes.

PAGURIDEA

Family Coenobitidae Dana
44. Ccenobita diogenes (Latreille)

Pagurus diogenes Latreille, Encycl. pi. 284, fig. 2 and 3.
Milne-Edwards, Hist. Nat. Crust., II, p. 240, PL 22, 1837.
{a) 2 $ . Nicol's Town, Andros Island, March 23, 1890.
{h) 2 $ , I 9 . Nassau, N.P., Jan. 16, 1890.

(d) 1 9 , juv. In pools and under stones. New Providence and
neighboring cays.

82 CRUSTACEA FROM THE BAHAMAS

Range : Florida to Brazil, West Indies, Bermuda.
Collected at Antilles, Cuba, Jamaica, Hayti, Turks Island, St.
Thomas, Barbadoes.

Family Paguridae

45. Petrochirus granulatus (Olivier)

Pagurus granulatus Olivier, Encyc. Meth., VIII, p. 640, 181 1.
Henderson, in Challenger, Anomura, p. 58, 1888.

(a) 3 $ . Largest about 8 in. long.

(b) I $ , I 9 ' In shell of Strombus gigas, Nassau, N.P., Jan. 26,
1890.

Range: West Indies, Gulf of Mexico to Brazil, Cape of Good
Hope.

Collected at Antilles, Cuba, Jamaica.

The common large West Indian hermit crab.

46 (?). Clibanarius vittatus (Bosc.)

Pagurus vittatus Bosc, Hist, des Crust., II, p. 8, pi. XII, 1802.
Kingsley, Proc. Acad. Nat. Sci. Phil., p. 236, 1878.

(a) I $ , imperfect. In small shell of Strombus gigas, beach near
Nassau, N. P., Jan., 1890.

(b) I (?), fragmentary. Near Nassau, N.P., Feb. i, 1890.
Range : Ft. Macon to Florida, West Indies, Brazil.

I refer these imperfect specimens doubtfully to this species. The
chelie are wanting in (a) , and (b) is too much broken to be of any value
in the determination.

47 (?). Clibanarius tricolor (Gibbes)

Pagurus tricolor Gibbes, Proc. Amer. Assoc, p. 189, 1850.

(a) Several specimens.

(b) 1 $. South side New Providence, in small shells of Strombus
gigas.

The determination is doubtful, as the specimens are very poor
and have almost entirely lost their color. They are all withdrawn into
the shells of various littoral mollusks.

CRUSTACEA FROM THE BAHAMAS 83

Family Parapaguridas Smith
48. Parapagurus sp.

(a) 2 9. Dredged, Jan. 22, 1890, Nassau, N.P.
Length of thorax 3 and 5 mm. respectively.
I refer these imperfect, colorless specimens doubtfully to some
species of Parapagurus.

LORICATA

Family Palinuridae Bate
49. Palinurus argus (Latreille)

Palinurus argus Latr., Milne-Edwards, Hist. Nat. Crust., II, p.
300, 1837.

(a) I ^,1 ?. New Providence, Jan. 27, 1890. Holes in sand
between tides, about 5-6 in. deep, ''very shy." ^

(6) I ? . Nassau, N.P., Jan., 1890. (Dry.)

Range : West Indies to Brazil.

Collected at Antilles, Cuba, Jamaica.

STENOPIDEA

Family Stenopidae Bate
50. Stenopus hispidus (Latreille.) (PI. XXIX, fig. i.)

PalcBMon hispidus Olivier, EncycL, VIII, p. 666, 181 1.

Stenopus hispidus Latr., Regne animal de Cuvier, ed. 2, IV,

P.93-

Bate, Challenger, Macrura, p. 211, PI. XXX.

Herrick, The Life History of Stenopus, Nat. Acad, of Sciences,
Vol. V, p. 339.

(a) 1 $ . Nassau, N.P., Jan. 22, 1890. "In life the antennae are
carried in front, not bent back."

I note the characters of special importance in order to compare
this already described species of Stenopus with the two species follow-
ing. Rostrum with a median dorsal row of 6 spines bifurcated at
extremity, a lateral row of 3 or 4 spines on each side of rostrum ; no
ventral spines. Back of the sixth dorsal spine a double row. Rostrum

^ This label is marked as doubtfully belonging to this specimen.

84 CRUSTACEA FROM THE BAHAMAS

does not reach to end of peduncle of inner antennae. Carapace of thorax
very rough, with firm, sharp spines which are longer on the dorsal
than on the lateral regions. Abdomen thickly armed with outwardly
projecting spines. Third pereiopod long, abundantly armed with
spines. The propodos with six rows above and below and two on each
lateral surface.

Measurements: total length 50.5 mm., length of cephalo-thorax
16.5 mm., of abdomen 34 mm., of rostrum 6 mm., of telson 9.5 mm.

Unless the eastern form should prove distinct from the West
Indian, we have a widely distributed species occurring in the warm wa-
ters of both hemispheres. It has been reported from : Indian Ocean
(Olivier), Australia (Peron and Lesneur), Borneo and Philippines
(Adams and White), South Pacific (Dana), Ambonia (DeMan),
Fiji Islands and Bermuda (Bate), Cuba (Von Martens), Bahama
Islands (Herrick).

I introduce a figure of this specimen (PL xxix. Fig. i), although not
a new species, in order to compare it with the two following species,
figures of which have not yet appeared.

51. Stenopus semilaevis Von Martens (PL xxix, Fig. 2)

Von Martens, Ueber Cubanische Crustaceen, Arch. f. Naturgesch.,
Bd. 38, p. 144, 1872.

(a) I $, 1 9 with ova. Under large sponge. New Providence,
Jan., 1890.

My specimens correspond very closely, except in certain minor
particulars noted below, with the description given by Von Martens of
a species "probably from the West Indies," which he found unde-
scribed in the Berlin Museum and which he called S. semilcEvis.

Von Martens' description (I.e. supra) I reproduce: "Cephalo-
thorax spiny; abdomen smooth; rostrum short, not longer than the
peduncle of the inner antennae, compressed laterally and prolonged as
a ridge nearly to the sharply marked cervical furrow, above with four
teeth, below teeth wanting. Carpus of third pair of pereiopods quad-
rangular as in S. hispidus, but the chelae compressed, with smooth
sides and not so long; chelae, including the dactyl, twice as long as
broad ; the upper margin sharper than the under and smooth, the under
serrated. The dactyl half as long as the palma; the back of the dactyl
keeled, serrated. Length from tip of rostrum to tip of telson 12 mm.

CRUSTACEA FROM THE BAHAMAS 85

Length of third pereiopod 13 mm. Breadth of chela 3 mm. The
fourth pereiopod shorter than third."

I note the following peculiarities in my specimens : Dorsal surface
of rostrum with six teeth; the fourth and sixth have each a minute sub-
sidiary tooth. Ventral surface with a single, not very prominent tooth.
Both margins of the chelae of the third pereiopods very finely serrated,
a'^rather prominent keel on the upper margin. The third pereiopod of
the right and left sides similar. Telson spiny. The large specimen
(?) is 15 mm. long, the Z slightly smaller. Length of chelae in ?
6 mm., breadth 2.5 mm.

Not having the opportunity of comparing the Bahama specimens
with Von Martens' type I prefer to consider these slight variations as
possibly due to imperfect description, and to place my specimens,
provisionally, at least, with Von Martens' species.

S. semilcBvis differs from 5. hispidus mainly in the teeth of rostrum,
the shorter rostrum, the proportionately shorter and thicker hand, the
less spiny carapace of cephalo-thorax and the smooth abdomen.

52. Stenopus scutellatus n. sp. (PL xxix, Fig. 3)

(a) I ( ?) Z . Under coral, near low water. Silver Cay, N.P.

Total length from tip of rostrum to tip of thorax 18 mm. Length
of rostrum 3 mm., of cephalo-thorax 7 mm.

Rostrum has a single row of ten spines on median dorsal line ; back
of the tenth a double row of three spines extend to the cervical furrow.
On median ventral line of rostrum are six spines ; no lateral spines on
rostrum. Rostrum longer than in S. hispidus, extending beyond the
peduncle of inner antennae. Whole surface of carapace covered with
delicate spines obscurely arranged in rows; usually curved forward,
with a somewhat reflexed tip. Spines on dorsal surface of first two
abdominal segments short and straight in a double row pointing for-
wards; on third segment several rows, stouter, pointing outwards;
on the fourth, fifth, and sixth segments spines are longer, pointing back-
wards. In the middle of the posterior portion of the tergum of the
third abdominal segment there is a polished, slightly elevated, shield-
shaped area, with crenellated margins, about i mm. in length. The
median tergal region of fourth segment is smooth and polished, sur-
rounded by a row of appressed spines, the same being true to a less
extent of the fifth segment. I have taken the specific name from this

86 CRUSTACEA FROM THE BAHAMAS

peculiar scutellar area on the third abdominal segment. This feature
seems to correspond to a triangular but less prominent area on the
similar segment in S. hispidus which is prolonged into a smooth dorsal
ridge on the next segment.

Telson lance-shaped, with a double row of spines between which is
a longitudinal groove about as long as the uropodal lamellae, which are
finely, serrated on their margins, and, as the telson, fringed with stiff
hairs.

Eyes on short peduncles which are armed above with three short
spines projecting over the cornea, and with a few spines at the anterior
margin. Cornea (in alcoholic specimen) bluish black. Inner anten-
nae ; peduncle with a few weak spines at distal end of segments. Outer
antennae; peduncles with strong, forwardly projecting spines. Scale
lined on inner margin with long closely set hairs and prolonged into a
ciliated bristle. Flagella more than twice the length of body. Third
maxillipedes, when extended, reach a little farther than extremity of
rostrum; the three distal segments about equal in length.

First pair pereiopods wanting in my specimen. Second pair
slender, chelate, segments of equal length. Third pair of similar pro-
portions to those in S. hispidus; chelae 7 mm. long; propodos laterally
compressed and somewhat triangular in cross section, broad above;
on the dorsal margin a double row of eleven spines each, on the ventral
margin a single row of nine spines ; two or three rows of minute spines
on lateral surfaces. A number of long soft hairs over the fingers,
especially at the tips. Hands of the two chelapods similar in size. Car-
pus and ischium together about equal to propodos, each armed with
rows of spines. Fourth pair long and slender ; dactylus bifid ; propodos
slightly spiny, one-half length of carpus. Carpus and propodos ob-
scurely articulated. Fifth pair pereiopods undeveloped. Pleopods
biramous, except first, with two or three spines each on the protopodite.

From the single specimen at my disposal I would compare this
species with S. hispidus as follows : rostrum proportionately longer,
nearly one-half length of cephalo-thorax, in n. sp. (J in hispidus), longer
than peduncle of inner antennae. Six ventral teeth (hispidus none), no
lateral teefh, single dorsal row of ten teeth (hispidus six). Flagella
of outer antennae fully twice the length of body ; proportion 2 : i f or
n. sp., 7:5 for hispidus. Tergum of third abdominal segment with
shield-shaped area. Third maxillipedes proportionately shorter than

CRUSTACEA FROM THE BAHAMAS 87

in hisptdus. Spines on cephalo-thorax equally long, but less rigid than
in hispidus, giving in general a less thorny character to the new species.

EUCIPHIDEA

Family Palaemonidae Bate
53. Palaemon savignyi Bate

Brachycarpus savignyi Bate, Challenger, Macrura, p. 795, PL
129, fig. 4, 1888.

Ortmann, Zool. Jahrb., Bd. V, p. 727.

{a) I specimen. Near Nassau, N.P., Feb., 1890.

{h) I specimen. Nassau, N.P., 1890.

{c) 5 ? with ova. Nassau, N.P., 1890.

Bate's specimen was from Bermuda, ''in shallow water."

"This is the most northern limit of genus Palaemon." Ortmann.

The species has not been described from any other localities.

54. Leander northropi n. sp. (PI. xxx. Fig. 4)

{a) I specimen. Nassau, N.P., Jan., 1890.

A single specimen with a total length of 30 mm. Length of cephalo-
thorax to tip of rostrum 11.5 mm.

Cephalo-thorax with small tooth below orbit and a very minute tooth
below this and a little back from the anterior margin on the lateral
surface.

Length of rostrum to posterior end of orbit 7 mm., slightly curved
upwards toward apex. Ten teeth above, /owr below; the first dorsal
tooth forms with the tip of rostrum a bifid extremity. A long interval
between first and second tooth ; interval between second and third one-
third the length of that between first and second ; fourth, fifth, and sixth
teeth follow at slightly diminishing intervals, the sixth being over the
posterior part of orbit of eye. Seventh, eighth, and ninth teeth close
together, posterior to orbit of eye.

The first ventral tooth is a little in front of second dorsal, second
ventral below second dorsal; third and fourth at equal intervals be-
tween second ventral and orbit of eye.

Inner antennae : peduncle reaches beyond second ventral tooth of
rostrum ; proximal segment about equal to the two distal. Upper flagel-
lum bifid; united proximal portion of 14 segments; the shorter branch

CRUSTACEA FROM THE BAHAMAS

has 12 segments ; united therefore for more than half its length. The
longer branch reaches beyond the undivided flagellum.

Outer antennae : scaphocerite with lamellar portion slightly longer
than spinose, reaches beyond first ventral tooth of rostrum ; flagellum
exceeds the length of the body.

Third pair maxillipedes reach to end of peduncle of inner antennae.

First and second pereiopods : long, slender, and chelate ; second
longer than first ; chela in second as long as carpus. Third and fourth
pereiopods terminate in claws.

Pleopods biramous, setose. Telson lanceolate, 4 mm. long, notice-
ably shorter than uropods, distal extremity with two sharp spines.
Outer uropod imperfectly divided transversely, the proximal division
ending in a lateral spine.

This species is allied to L. petitinga F. Miiller, from Brazil (see
Ortmann, Revista do Museu Paulista, II, p. 191, 1897) and to L.
maculatus Thallwitz (Abh. Mus. Dresd., Ill, p. 19, 1891) from West
Africa.

I am indebted to Dr. Ortmann for the preparation of the following
table, which exhibits the relationship :

L. maculatus

T - f 12-13 segments united

Inner antennae-^ ^ °

(8 segments free.

6 + 1 f I posterior to

orbit.

Teeth of rostrum

+ 1 (" I

3 1

L. northropi

114 segments united
12 segments free.
9 + 1(4 posterior
4 [ to orbit.

L. petitinga
1 9 segments united
[ 20 segments free.
6 + 1 r I posterior
5 — 6 1 to orbit.

Family Hippolytidae Ortmann
55. Tozeuma carolinense Kingsley

Kingsley, Proc. Acad. Nat. Sci. Phila., p. 90, 1878.

(a) I <?, with ova. Dredged in about 16 ft. Near Quarantine
station, Jan., 1890.

Kingsley 's specimens are from Ft. Macon and Beaufort, N.C.,
and Charlotte Harbor, Fla.

Measurements of Bahama specimen: total length 41 mm., rostrum
12 mm., cephalo-thorax (without rostrum) 7 mm., abdomen 22 mm.

CRUSTACEA FROM THE BAHAMAS 89

Family Alpheidae Bate
56. Alpheus edwardsii Audouin

Athanas edwardsii Audouin, Planches de la descrip. de I'Egypte
par M. Savigny, Crust., PI. X, fig. 11, 1810.

Bate, Challenger, Macrura, p. 542, 1888.

{a) 4 specimens. Near Nassau, N.P., along shore, Feb. 20, 1890.

{b) I specimen. Nassau, N.P., Jan., 1890.

{c) 3 specimens. Under coral and in pools between tides, New
Providence.

{d) I specimen. Under coral and in pools between tides, Nassau,
N.P., Jan., 1890.

{e) 2 broken. Near Nassau, N.P., Feb., 1890.

The distribution of this species is circumtropical.

57. Alpheus hippothoe De Man

bahamensis, n. var. (PI. xxx. Fig. 5)

{a) 24 specimens. Under coral and in pools between tides, New
Providence.

(6) 3 specimens, one with ova. Nassau, N.P., Jan., 1890.

This species is most closely allied to the variety edamensis of
Alpheus hippothoe De Man, from the Bay of Bengal and Indian
Archipelago (Arch, de Naturg., Bd. 53, p. 518, 1887). I am indebted
to Dr. Ortmann for a communication from Professor de Man com-
paring specimens from my material with his own hippothoe. As there
are certain differences between the West and East Indian specimens,
I propose to make a new variety for the West Indian.

Total length from rostrum to telson, largest 24 mm., smallest 15
mm. Rostrum reaches nearly to end of first segment of inner antenna,
sharp, laterally compressed, prolonged backwards as a distinct keel.
Between keel and the prominent eyes a rounded depression. No ocular
spines.

Inner antennae : first joint of peduncle with small spine on outer
surface; second joint nearly twice the length of proximal; terminal
joint one-half the length of second. Shorter flagellum, about the
length of peduncle. Longer flagellum slender, about thrice the length
of shorter.

90 CRUSTACEA FROM THE BAHAMAS

Outer antennae : peduncle a little longer than that of inner, small
spine on basal joint. Flagellum one-third longer than long ramus of
inner antenna, spinose portion of scaphocerite a little longer than the
peduncle. Flabellar portion (scale) a trifle shorter ; not quite so long
as the peduncle.

Third pair of maxillipedes do not reach beyond end of peduncle of
the outer antennae.

First pair pereiopods: large chela of largest specimen has a
length of 1 8 mm., of smallest specimen 8 mm. The large chela has a
somewhat quadrangular depression on the outer surface, the distal
end of which is continued upwards into a well-marked depression on
the dorsal margin and extends backward as a groove along the inside
of the dorsal surface. A distinct but less marked depression on the
ventral margin. Inner surface of the hand slightly hairy, outer surface
nearly smooth. Fingers contorted, color in alcoholic material pale
blue. Movable finger slightly longer than thumb. In the small
chelapod, which may be on the right or leftside, the finger is
one-third the length of palm. Carpus of chelapods short. Meros
triangular in section; ends distally in a sharp spine on the outer and
inner angle. Distal end of meros reaches to end of peduncle of outer
antennae.

Second pair of pereiopods very long. Distal end of meros reaches
beyond antennal peduncle. First and second joints of carpus sub-
equal, each a little longer than third and fourth together. Third and
fourth subequal. Fifth about two-thirds length of first ; equal in length
to fourth and fifth together. Finger about one-half length of thumb.
(Fifth joint a little too short in figure.)

Third and fourth pereiopods short and stout, not quite reaching to
distal end of meros of second. Length of meros less than three times
its breadth. Carpus one-half length of meros. Both carpus and meros,
with spines on lower margin of distal end. Propodos serrated on pos-
terior surface.

Fifth pair of pereiopods shorter and more slender. Telson with
median furrow. Two small spines on either side of furrow. Outer
plate of uropod minutely serrated on end. A sharp spine on its outer
distal angle.

Principal variations from A. hippothoe:

In new variety: peduncle of outer antennae longer than that of

CRUSTACEA FROM THE BAHAMAS 91

inner. Lamellar portion does not reach end of peduncle. Third
maxillipedes do not reach beyond antennal peduncle. Relative
lengths of carpal joints of second pereiopods differ.

Variations from var. edamensis :

Finger of small hand shorter than palm (longer in edamensis).
A quadrangular rather than a triangular depression on side of large
hand.

First joint of carpus of second pereiopod is equal in length to second
(shorter in edamensis). Third and fourth pereiopods less broad than
in edamensis.

58. Alpheus Websteri Kingsley

Kingsley, Proc. Acad. Nat. Sci. Phila., p. 416, 1879.

(a) 3 specimens, one with ova. Along shore, near Nassau, N.P.,
Feb. 20, 1890.

(b) 2 specimens, one with ova. Nassau, N.P., Jan. 10, 1890.

(c) 3 specimens, fragmentary. Nassau, N.P., Jan. 10, 1890.
Kingsley's type specimens were from Key West; it has been re-
ported by Herrick from Nassau, N.P.

A. Websteri is very probably the same as A. formosus Gibbes
(Proc. Am. Ass. Ad. Sci., p. 196, 1850). The descriptions apparently
tally, though Gibbes makes no mention of the small black spine on the
uropod which is mentioned as a characteristic feature by Kingsley,
and which is very evident in my specimens.

59. Alpheus nigro-spinatus n. sp. (PL xxx. Fig. 6)

(a) 2 specimens. Under coral and in pools between tides. New
Providence.

Carapace compressed. Rostrum short, acuminate, no longer than
spines of ocular hoods; extended backwards as a ridge between the
eyes, from each of which it is separated by a rounded depression.
Spines of ocular hoods short, acuminate. The front of carapace is
thus marked by three, nearly equal, small spines. Inner antennae:
basal segment of peduncle with small spine (stylocerite) ; second and
third segments, no spines but scattered hairs ; second segment a little
more than twice as long as the terminal ; outer flagellum stouter and
shorter than the inner. Outer antennae: outer angle of the basal
joint of peduncle with a sharp short spine; scaphocerite broad at

92 CRUSTACEA FROM THE BAHAMAS

base, outer margin produced into a strong spine which is longer than
the inner, lamellar, portion; distal end of terminal segment of peduncle
reaches to tip of scaphocerite. Third pair of maxillipedes reach about
to the end of shorter flagellum of inner antennae ; strongly tufted with
hair.

First pair of pereiopods : larger hand much inflated, a slight but
distinct constriction on the upper margin near the articulation of the
dactylus, and a deep constriction on the lower margin. Thumb con-
torted ; a groove on the outer margin, the inner surface thickly covered
with hairs and punctate. Dactylus contorted ; extends slightly beyond
thumb ; inner surface with tufts of hair. Small hand (which on the one
specimen is left, the other right) has a longer and more slender dactylus
and thumb. Length of large hand i6 mm., breadth 6.5 mm. Length
of small hand 9 mm., breadth 4 mm.

Second pair of pereiopods : carpus five-jointed, proximal segment
the longest, slightly longer than the second and third together ; second
and fifth segments each a little longer than one-half the length of first ;
third and fourth the shortest, subequal. Posterior pereiopods:
meros without spines. Telson broadly triangular ; extremity truncate ;
two small spines on either side of median line of dorsal surface; the
outer ramus of uropod bears on its external distal angle a large, very
black spine, which is distinguished from the similar black spine of A.
Websleri Kingsley (I.e., p. 416, 1879) by its much larger size and con-
sequently more prominent appearance. Length of specimens 25 mm.
and 22 mm. respectively.

60. Alpheus minor Say

Say, Jour. Acad. Nat. Sci. Phil., I, p. 245, 1818.
Kingsley, Bull. U. S. Geol. Survey, IV, p. 190, 1878.
Bate, Challenger, Macrura, p. 558, PL C, 1888.

(a) numerous specimens, from brown sponges.

(b) 1 9, with ova. Along shore near Nassau, N.P., Feb. 20, 1890.

Range: From Cape Hatteras (U.S.F.C. 1885) to St. Paul's
Rock (Bate, Challenger). Both shores of Central America.

Collected at Jamaica, New Providence.

Lot (c) may possibly be a variety, as the thumb is shorter than the
typical minor, but otherwise there seems no difference.

CRUSTACEA FROM THE BAHAMAS

6i. Alpheus saulcyi Guerin

Guerin, in Hist, du Cuba, 1857.

Herrick, Memoirs Nat. Acad. Sci., Vol. V, p. 381.

{a) 5 specimens, from green sponges, Feb., 1890.

{h) I specimen, near Nassau, Feb. 5, 1890.

{d) I specimen, from sponge, March i, 1890.

(e) I specimen, from sponge, March i, 1890.

(/) 2 specimens, Nassau, N.P., Jan., 1890.

Range : West Indies.

Found at Nassau, Martinique.

62. Athanas ortmanni n. sp. (PI. xxx, Fig. 7)

(a) I specimen. Along shore, near Nassau, N.P., Feb. 20, 1890.

Rostrum slender and pointed, reaching a little beyond the second
joint of peduncle of inner antennae. Antero-lateral margin of carapace
extends obliquely backward, prolonged in front of eye into minute
spine. Eye-stalk short, not projecting beyond carapace. The eye is
seen through the somewhat transparent carapace as in Alpheus.
Inner antennae, with stylocerite reaching to distal end of second seg-
ment of peduncle. From the peduncle arise two flagella of nearly
equal length, the upper somewhat more slender than the lower, bearing
on the fourth segment from base a minute, subsidiary flagellum.

Outer antennae with scaphocerite nearly as long as the peduncles of
inner antennae, broad and fringed with hairs. Third pair of maxilli-
pedes reach slightly beyond the distal end of scaphocerite.

First pair of pereiopods: that on the right side is robust with
swollen chela, terminating in slender hooked fingers which are minutely
serrated on the opposing edges. Margin of chela entire, length 5 mm.,
breadth 2.5 mm. Carpus short. Distal end of meros reaches to
extremity of third pair maxillipedes. Left chelapod lacking.

Second pair of pereiopods slender, with very small chelae. Carpus
five-jointed; proximal segment equal in length to the four distal seg-
ments. Remaining three pairs of pereiopods similar to each other and
equal in length to the second pair. Pleopods narrow and biramous.
Telson narrow and compressed, with smooth margins. Uropods
slightly longer than telson.

94 CRUSTACEA FROM THE BAHAMAS

Total length of specimen i6 mm.

The species above described agrees generically with Athanas
Leach (Edin. Ency., VIII, p. 432), with the exception that the eyes
are entirely covered by the carapace. I propose, rather than found a
new genus on the single specimen, to amend Leach's definition of
Athanas by changing the statement, " Opthalmopoda short, scarcely
reaching beyond frontal margin of carapace" (Bate, Challenger,
Macrura, p. 528), to opthalmopoda short, covered by, or scarcely reach-
ing beyond the frontal margin of carapace.

There are four hitherto described species of Athanas : —

A. nitiscens Leach. England and Norway, Mediterranean to
Cape Verde Islands.

A. veloculus Bate (I.e., p. 529). Cape Verde Islands.

A. mascarenicus Richters (Beitrage zur Merresfauna von Mauri-
tius u. d. Seychellen, p. 164, 1880), Mauritius.

A. dimorphus Ortmann, Crust, in Semon's Forschungsreise (Jena.
Denks., VIII, 1894, p. 12). East Africa: Dar-es-Salaam.

From all these species A. ortmanni may be distinguished at a glance
by the form of the large chela.

PENiEIDEA

Family Penaeidae Bate

63. Penaeus constrictus Stimpson

Stimpson, Ann. Lye. Nat. Hist. N.Y., p. 135, 1871.
Miers, Notes on the Penaeidae, Proc. Zool. Soc, London, p. 308,
1878.

(a) I Z. Near Nassau, N.P., Feb. i, 1890.

(6) I ?. Nassau, N.P., Feb. 5, 1890.

Range: East coast United States.

Not before reported from West Indies.

Collected by Stimpson at Beaufort, and Charleston, S.C.

STOMATOPODA

Family Squillidae Latreille

64. Pseudosquilla ciliata Miers

Miers, Ann. and Mag. Nat. Hist. (5) V, p. 108, PI. Ill, figs. 7 and 8,
1880.

CRUSTACEA FROM THE BAHAMAS 95

Brooks, Challenger, Stomatopoda, p. 53, 1886.
(a) I ^ , broken. Near Nassau, N.P., Feb., 1890.
Range : Widely distributed over Atlantic and Pacific.
Collected at Cuba, Bahamas, St. Thomas.

65. Gonodactylus oerstedii Hansen

Hansen, Isopoden, Cumaceen und Stomatopoden der Plankton
expedition, 1895.

(a) I 9. Nassau, N.P., Feb. 5, 1890.

(b) I 9 , fragmentary. Quarantine station, near New Providence,
Jan. 25, 1890.

(d) I ?. Nassau, N.P., Jan., 1890.

(e) I ^ , I 9 , I fragmentary. Under coral and in pools between
tides, near Nassau, N.P.

(J) I $. (label erased).

(g) 2 juv. Dredged in about 16 ft. near Quarantine station, Jan.,
1890.

Hansen, I.e. supra, p. 65 (and footnote), calls the West Indian
Gonodactylus: G. oerstedii n. sp. and retains the name G. chiragra
Fabr. for the East Indian form.

He says (footnote) : *'This species {cerstedii) maybe distinguished
from the East Indian form, G. chiragra Fabr., especially by the char-
acter, that it possesses a small keel inside of and close to the keel that
ends in the sublateral process of the posterior margin, while such a
secondary keel is wanting in the Indo-Australian species."

Collected at Bahamas, Cuba, Jamaica, St. Thomas.

CIRRIPEDEA

Family Lepadidae Darwin

66. Lithotrya dorsalis Sowerby

Sowerby, Genera of shells, April, 1822.

Darwin, A Monograph of the Cirripedia, p. 351, PL VIII, fig. i,
a, 1851.

(a) 10 specimens. Salt Cay, N.P., in rocks in surf, Jan. 28, 1890.
{h) 8 specimens. Salt Cay, Nassau, N.P., ocean side, Feb. 6, 1890.

96 CRUSTACEA FROM THE BAHAMAS

Range : West Indies, Venezuela, Honduras.
Collected at Barbadoes.

Family Balanidae Darwin
67. Acasta cyathus Darwin

Darwin, A Monograph of the Cirripedia-Balanidae, p. 312, PL IX,
1854.

(a) 4 specimens, in sponge. Dredged Jan. 22, 1890.

(b) 2 specimens. Near Nassau, N.P., Feb., 1890.
Range: Madeira, West Indies (Darwin).

ISOPODA

Two species of Isopoda, one probably a Lygia of which there are
several specimens. Another parasitic on a fish, probably one of the
CirolanidcB.

AMPHIPODA

Several small amphipods undetermined.

PLATE XXIX

RWeOer, cfel.

Fig. I. — Stenopus hispidus Latreille.

Fig. 2. — Stenopus semilaevis Von Martens.

Ji.Weberj del..

Fig. 3. — stenopus scutellatus, n. sp. Rankin.
97

PLATE XXX

Fig. 4. — Leander northropi, n. sp., Rankin.

Fig. 5. — Alpheus hippothoe De Man.

var. bahamensis, n. var., Rankin.

Fig. 6. — Alpheus nigro-spinatus, n. sp., Rankin.

R.mber, del.

Fig. 7. — Athanas ortmanni, n. sp., Rankin.
98

LIST OF SHELLS COLLECTED BY DR. JOHN L

NORTHROP IN THE BAHAMAS, IDENTIFIED

BY PROFESSOR WILLIAM HEALEY DALL,

SMITHSONIAN INSTITUTION

The following list was made from Dr. Northrop's specimens which
were sent to the National Museum for the purpose. So many shells
have been collected at the Bahamas during the last two hundred years
that the fauna is fairly well known, and it was not to be expected that
any very remarkable additions to it could be made on such a collect-
ing trip. Well-identified species, however, always have a certain value
for science, in their bearing on geographical distribution, as the locali-
ties given by the earlier collectors were often vague or inaccurate.

That Cyrena floridana and Macroceramus pontificus should occur
was interesting, and a number of the other species, though known to
belong to the general region, had not heretofore been specifically
located in the Bahamas.

The shell identified as Mitra flammea is bleached and worn, and
may be another species. It is not, however, any of those commonly
known as Antillean.

PELECYPODA

138. Area occidentalis Philippi. Deep Creek, Andros.

103. Arcafusca Brug. Salt Cay.

35. Area jamaicensis Gm, Lake Cunningham, N. P.

39. 210.* Glycymeris undulata Lam. Salt Cay.

16. Pinna carnea Gm. Nassau.

92. Melina ohliqua Lam? Salt Cay.

46. Melina ephippium Linn. Salt Cay.
134. Pteria atlantiea Linn. Deep Creek, Andros.

59. Margaritiphora radiata Leach. Salt Cay.
136. Ostrea frons Linn. Deep Creek, Andros.

65. Peeten ornatus Lam. Salt Cay.
60; 137. Spondylus spathtdiferus Sby. Salt Cay and Deep Creek, Andros.
123. Lima scabra Born. Nassau.

104. Lithophaga antillarum Phil. Salt Cay.

102; 130; 224. Mytilus exustus Linn. Salt Cay.

100 LIST OF SHELLS COLLECTED IN THE BAHAMAS

151.
125.
72;

76.
44.
48.
66.

51;
40;

43.

99.

128;

41.

17-

74.

46a.

45-

75-

Cyrena colorata Prime. Cedar Coppet, west side of Andros.

Cyrena Jloridana Con. Low Sound, north end of Andros.
129. Chama macerophylla Chemn. Salt Cay, N. P., and Deep Creek, east side
of Andros.

Chama sarda Rve. Salt Cay.

Phacoides pensylvanicus Linn. Salt Cay.

Codakia Linn. Beach near Nassau.

Divaricella quadrisulcata Orb. Salt Cay.

96. Cardium serratum Linn. Salt Cay.

222. Cardium medium Linn. Salt Cay.

Cardium isocardia Linn. Salt Cay.

Anomalocardia cuneimeris Conr. Salt Cay.

133. Chione cancellata Gm. Wide Opening, west side of Andros and Deep
Creek, Andros.

Chione paphia Linn. Salt Cay.

Asaphis deflorata Linn. Near Deep Creek, Andros.

Tellina fausta Don. Salt Cay.

Tellina interrupta Wood. Beach near Nassau.

Tellina magna Spengl. Salt Cay.

Semele reticulata Gm. Nassau.

TERRESTRIAL GASTROPODA

121. Polygyra cereolus Muhl. Red Bays, west side of Andros.

119. Cepolis albersiana Pfr. Red Bays, west side of Andros.

122. Cepolis Duclosiana Fer. Red Bays, west side of Andros.

118. Cepolis varians Mke. Red Bays, west side of Andros.

120. Cepolis troscheli Pfr. Red Bays, west side of Andros.
18; 80. Pleurodonte auricoma Fer.. Nassau.

228.* Orthalicus undatus Brug.
112. Tralia pusilla Gm. Salt Cay.

155. Melampus bidentatus Say. Cedar Coppet, west side of Andros.
62 ; 223.* Melampus coffeus Linn. Salt Cay.
28. Opisthosiphon Bahamense. Nassau.

2. Cerion (Strophiops) Sagraianum Pfr.
183-184. Cerion (Strophiops) glans Kiister., and var.

119. Cerion (Strophiops) injlatum Mayn.

3. Cerion (Strophiops) glans, var. cinerea Mayn.

4. Cerion (Strophiops) cyclostomum Kiister.

5. Cerion (Strophiops) lentiginosum Mayn.
6; 7; 8. Cerion (Strophiops) maritimum Pfr.

MARINE GASTROPODA

150. Bullaria striata Brug. Cedar Coppet, west side of Andros.
58. Bullaria occidentalis A. Ad. Salt Cay.

LIST OF SHELLS COLLECTED IN THE BAHAMAS 101

53. Terehra casta Hds. Salt Cay.
157a. Terehra protexta Con. Cedar Coppet, west side of Andros. ^

6. Conus mus Hwass. Salt Cay.
4; 105a. Thais patula Linn, {operculum). Beach near Nassau. Salt Cay.
91. Thais hcBmastoma Linn. Salt Cay.
47. Thais deltoidea Lam. Salt Cay.
85. Leucozonia ocellata Gm. Salt Cay.
85a. Leucozonia leucozonalis Lam. Salt Cay.
9. Turbinella scolymus Gm. Nassau.
8. Fasciolaria tulipa Linn. Nassau.
^ I. Oliva reticularis Lam. Nicholas Town, Andros.
36; 95. Columbella mercatoria Linn. Salt Cay.

63. Nitidella nitida Lam. Salt Cay.
94. Mitra Barbadensis Gm. Salt Cay.

162. Mitra Jlammea Quoy. ? Salt Cay.

12. Marginella guttata Dill. Near Nichol's Town, Andros.

113. Marginella lactea Kiener. Salt Cay.

124. Marginella apicina Mke. Low Sound, north end of Andros.

154. Marginella apicina Mke. ? Cedar Coppet, west side of Andros.

89. Cymaiium pileare Linn. (juv.). Salt Cay.

III. Septa tritonis L., var, nobilis Con. Stafford Creek, Andros.

22. Cassis testiculus Linn. Salt Cay.

21. Cassis injlata Shaw. Salt Cay.

105. Morum oniscus Linn. Salt Cay.

no. Tonna perdix Linn. Nichol's Town, Andros.

3c.* Cyphoma gibbosa Linn.

3. Cyprcea spurca Linn. ? Salt Cay.

3a. Cyprcea cinerea Linn. Salt Cay.

7. Cyprcea exanthema Linn. Mastic Point, Andros.

64. Trivia pediculus Linn. Salt Cay.

31. Cerithidea costata Wood. Border of pond. Salt Cay, near Nassau.
162a. Cerithium literatum Born. Salt Cay.

157. Cerithium uncinatum Gm.? Cedar Coppet, west side of Andros.
127. Vermetus radicula Stimp. Wide Opening, west side of Andros.
55 ; 126. Littorina angulifera Lam. Mangroves, Mastic Point, Andros and Nas-
sau.

25. Littorina ziczac Dillw. Tide pools, Nassau.

30. Littorina mespilum Mke. Tide pools, Salt Cay.
107; 116. Tectarius muricatus Linn. Nassau; Salt Cay.
49; 117. Echinella nodulosa Pfr. High tide, Nassau.

87. Hipponix antiquatus Linn. Salt Cay.

13; 32; 69; 149. Natica canrena Linn. Near Nichol's Town, Andros; Salt
Cay, Nassau; Cedar Coppet, west side of Andros.

24. Acmcea melanoleuca Gm. Rocks between tides, Nassau.

67. Acmcea puncturata Lam. Salt Cay.

102 LIST OF SHELLS COLLECTED IN THE BA.HAMAS

15. AstrcBa ccBlata Chemn. Cays near Nassau.
81; 227.* Astrcea longispina Liaim. Salt Cay.

93. Tegula scalaris Anton. Salt Cay.
loi. Tegula fasciata Born. Salt Cay.
10; 208.* Livona pica Gm. Nassau.
68; 217.* Nerita tessellata Gm. Tide pools, Salt Cay.
212.* Nerita versicolor Lam.
213.* Nerita peloronta Linn.
164. Neritina virginea Linn, Andros ( ?) .

29. Neritina pupa Linn. Tide pools, Salt Cay.

88. Fissurella alternata Say. Salt Cay.
2. Fissurella Listeri Orb. Salt Cay.

2a. Fissurella Barbadensis Gm. Salt Cay.

CEPHALOPODA

37. Spirula australis Lam. Gulfweed on beach, Nassau.

N.B. — The numbers with an asterisk had no localities given for them.

NOTES ON SOME ACTINIANS FROM THE BAHAMA
ISLANDS, COLLECTED BY THE LATE DR. J. I.
NORTHROP '

By J. Playfair McMurrich, Ph.D.

(Read April ii, 1896.)

The Actinians which form the subject of this paper were collected
by the late Dr. J. I. Northrop, of Columbia College, during a visit to
the Bahama Islands some years ago. It had been Dr. Northrop's
intention to work over the collection himself. Before his lamented
death he had studied several species more or less thoroughly. The
drawings and preparations which he had made were intrusted to me,
together with the collection, by Mrs. Northrop, and have proved of
great assistance in determining the affinities of certain forms. Espe-
cially was this true in regard to Heteractis lucida, whose systematic
position Dr. Northrop had correctly determined, and the notes on
Diplactis bermudensis and Isaurus duchassaingi were also very full.

I have taken pains to verify all the important facts contained in the
notes, and it has seemed more convenient for the present purpose to
describe in my own words the various forms to be considered, rather
than to attempt a rearrangement of Dr. Northrop's notes with such
additions as might seem necessary.

The following is a list of the species in the collection.

HEXACTINI^
Suborder Actininae

Family SAGARTID^

1. Aiptasia tagetes (Duch. & Mich.) Andr.

2. Heteractis lucida Duch. & Mich.

Family ANTHEAD^

3. Condylactis passijiora Duch. & Mich.

1 Annals N. Y. Acad. Sci., Vol. IX, July, 1896. — 13.
103

104 ACTINIANS FROM THE BAHAMA ISLANDS

" ""^~ Family DENDROMELID^

4. Lebrunea neglecta Duch. & Mich.

Family PHYLLACTID.E

5. Diplactis bermudensis McMurr.

Suborder Stichodactylinae

Family DISCOSOMID^

6. Discosoma anemone (Ellis) Duch.

Family RHODACTIDiE

7. Rhodadis sancti-thomce (Duch. & Mich.) McMurr.

8. Ricordea florida Duch. & Mich.

Family PHYMANTHID^

9. Phymanthus crucifer (Lesueur) Andres.

ZOANTHE^

10. Zoanthus sociatus (Ells) Lesueur.

11. Zoanthus nymphcBUs Lesueur.

12. Isaurus duchassaingi Andres.

13. Palythoa flava (Lesueur) Duch. & Mich.

14. Gemmaria isolata McM.

Certain of these forms I have already described ('89), and conse-
quently shall not consider in detail here ; a few others require but brief
notice, while three others require a more thorough description. I shall
consider the forms belonging to the last two groups in the order in
which they are named in the above list.

Family SAGARTID^

Heteractis lucida Duch. & Mich.

Synonyms: Capnea lucida Duch. & Mich, i860 (PI. VI, Figs. 9 and 12).
Heteractis lucida Duch. 8z: Mich. 1866.
Ragactis lucida Andres. 1883.

Several specimens of this form were contained in the collection
which offered accordingly opportunities for an accurate determination
of its systematic affinities.

ACTINIANS FROM THE BAHAMA ISLANDS 105

In the contracted condition the animal is generally somewhat
flattened, measuring in height 1.5-2.0 cm., and in diameter about 2 cm.,
the base being usually a little smaller than the upper part of the col-
umn. Dr. Northrop's notes contain no statement as to the coloration
of the living animal, but according to Duchassaing and Michelotti the
mouth is white, surrounded by a band of brilliant yellow, and the ten-
tacles have the appearance of white glass.

The mesogloea of the column walls is rather thin, and in much con-
tracted forms may appear as if provided on its endodermal side vdth
numerous distinct processes, which give to the column wall a ridged
or checkered appearance. Toward the upper part of the column are
numerous cinclides, not readily distinguishable in all preserved speci-
mens and apparently without any very definite arrangements, though
they are confined entirely to the upper part of the column, none being
found below the middle.

There is a distinct collar and fosse in the preserved specimens just
above the cinclidal region of the column, and at that portion of the wall
which forms the floor of the fosse the endodermal muscle processes
are considerably higher than elsewhere in the column wall, and form
what may be termed a diffuse endodermal sphincter. The true
sphincter lies, however, higher up, just below the point where the tenta-
cles arise, and is a weak structure embedded in the mesogloea, which is
hardly appreciably thickened for its reception. It consists of a few
rather scattered cavities, arranged practically in a single row and con-
taining the remains of muscle cells; it is not strong enough to produce
complete inclosure of the tentacles.

On account of the position of the sphincter the collar cannot be
regarded as the margin. Indeed there is no distinct margin, the upper
part of the column wall passing directly over into the bases of the
tentacles, so that, in the usual formula, it is necessary to say that the
margin is tentaculate. The tentacles are rather numerous, apparently
somewhere in the neighborhood of 192, though an accurate count is
difficult in the preserved specimens on account of the manner in which
the cycles are crowded together towards the margin. They vary some-
what in length in the various specimens I have examined, owing to dif-
ference of contraction, but the average may be put at i cm. The most
interesting feature of the tentacles is, however, the occurrence upon them
of a large number of spherical protuberances scattered over their sur-

106 ACTINIANS FROM THE BAHAMA ISLANDS

face without any apparent regularity, though more abundant upon the
oral than on the aboral surfaces. In section (pi., fig. i) these are seen
to be hollow outpushings of the wall of the tentacles. The wall of the
tentacle proper has a moderately thick mesogloea, the ectodermal and
endodermal muscle processes being rather low, the ectoderm possess-
ing but few nematocysts, and the endoderm being thin with only a few
Zooxanthellae. The wall of the protrusion differs not a little from this ;
thus the mesogloea is much thinner, the muscle processes are aborted,
the ectoderm is richly provided with nematocysts, so that the protru-
sions might be termed batteries of nematocysts, and the endoderm is
thick and abundantly packed with Zooxanthellag.

The tentacles leave, exposed a considerable portion of the disk,
perhaps about half its breadth, and this portion is smooth and slightly
concave. Its mesogloea is rather thin, and the muscle processes are but
moderate in length. No muscle cells are inclosed within the mesogloea.
The mouth is not elevated above the disk, and is almost circular in the
preserved specimens, the gonidial angles being but slightly marked.

In the single specimen in which I counted the mesenteries there were
forty-eight pairs, a number which agrees with that found by Dr.
Northrop in another specimen. Of these forty-eight pairs but six
are perfect, forming the first cycle; the six pairs of the second cycle
nearly equal the first cycle in size, but are imperfect ; the twelve of the
third cycle are considerably smaller, though still quite muscular; while
the twenty-four pairs of the fourth cycle, though extending some dis-
tance in from the body wall, -yet have their musculature but feebly
developed. Two pairs of directives are present and are attached to a
greater extent of the stomatodaeum than are the other mesenteries of
the first cycle. The longitudinal muscles are well developed and have
in section the form represented in fig. 2. The parieto-basilar muscles
are but feebly represented and do not require special description.
Numerous acontia occur, a fact that was discovered by Dr. Northrop.
None of the specimens examined possessed reproductive organs, so
that nothing can be said regarding the distribution of these organs.

There seems to be no room for doubt but that this form is identical
with that described by Duchassaing and Michelotti in their first paper
('60) as Capnea lucida though later ('66) assigned to the genus Heter-
actis, a genus established by Milne-Edwards for the reception of the
Actinia aurora of Quoy and Gaimard. Andres {'8^) has referred it to

ACTINIANS FROM THE BAHAMA ISLANDS 107

the genus Ragactis, which he established for a species, R. pulchra,
discovered by him in the Mediterranean. The distinction between
Ragactis and Heteractis concerns practically the tentacles only, the
evaginations upon these structures being in the former genus small
and appearing as tubercles, while in the latter they involve the entire
circumference of the tentacle, which thus becomes moniliform. How
far this distinction is valid can only be determined by further examina-
tion of the anatomy of the various species assigned to the two genera,
and for the present it seems to me more convenient to retain for the
species under discussion the name proposed by its sponsors.

An interesting fact which has developed from the examination of
H. lucida is that it is unquestionably a Sagartid. The occurrence of
acontia (to which my attention was first called by Dr. Northrop) and
of cinclides, the existence of only six perfect mesenteries and the meso-
gloeal sphincter point indubitably to this relationship, and it will be
interesting to know whether Ragactis pulchra and the other species
assigned to the genus Heteractis are likewise Sagartids. Andres has
grouped together in the family Heteractidae several genera, e.g. in
addition to the two already mentioned, Eloactis, Rhopalactis, and
Stauractis. Of these, as I have already suggested ('93), the genus
Eloactis, as judged by the species E. Mazeli, is probably with greater
propriety referable to the family Halcampidae, and I have now shown
that one at least of the members of the genus Ragactis must also be
transferred to another family. Families based upon mere external
characters have no firm foundation, tentacles may vary enormously
in shape, warts or tubercles may be absent or present within the limits
of the same family, and cases like the present show, if proof be still
wanting, that the criteria for a proper classification of the Actiniaria
are to be obtained only by the careful study of the internal anatomy of
the various forms.

We must add then, provisionally, to the genera included in the
family Sagartidas, the genus Heteractis, since one of the forms assigned
to it has proved to be a typical Sagartid. Whether, however, tlie type
species of the genus, H. aurora, will also prove to be a Sagartid, or
whether it will be necessary to remove H. lucida from the genus Heter-
actis, remains to be determined. And furthermore the affinities of
Ragactis pulchra must also be accurately studied before the proper
generic term for H. lucida can be finally determined.

108 ACTINIANS FROM THE BAHAMA ISLANDS

Family DENDROMELID^

In my earlier paper ('89) on West Indian Actiniaria I proposed a
new subtribe Dendromelinae for the reception of the peculiar form
Lebrunea neglecta discovered by Duchassaing and Michelotti ('60)
and proposed to place in it also the genus Ophiodiscus described by
R. Hertwig ('82). Since that time I have come to the conclusion that
I gave the group too high a mark and have suggested ('93) its reduction
to a family of the subtribe Actininae. With the two forms already men-
tioned it seems that a third should be associated, namely, the Hoplo-
phoria coralligens described by H. V. Wilson ('90), which, as I have
already pointed out elsewhere ('93), seems to be identical with the
Viatrix glohulifera described by Duchassaing and Michelotti ('60).

Lebrunea neglecta Duch. & Mich, i860

A single specimen of this form was in the collection, and being
unwilling to injure it, I have nothing to add to the description of the
structure given in my earlier paper. It may be stated, however, that
the number of pseudotentacles is in this individual six, just as they
were in the other specimens examined; one, however, is considerably
smaller than the others.

Family PHYLLACTID.E

Diplactis bermudensis McMurrich. 1889

This species I described ('89^^) originally from alcoholic specimens
from the Bermudas and was pleased to find it represented in the
present collection by several specimens, together with a drawing (here
reproduced in pi. xvii, fig. 3) and a description of the living animal.
I have nothing to add to my original account of the anatomical char-
acteristics of the species, but am able to complete it by abstracting
from Dr. Northrop's notes a description of the coloration and form of
the living animal.

The column is chocolate-brown in color, and when the animal is
expanded, has a greater diameter at the base and at the upper part
than in the middle of its length, the upper part forming a slight parapet
projecting above the level of the disk. The tentacles are brownish red
in color and the disk rufous, that is to say, somewhat paler than the
tentacles. The papillae which represent the fronds Dr. Northrop

ACTINIANS FROM THE BAHAMA ISLANDS 109

apparently overlooked, inasmuch as they would be completely hidden
by the expanded tentacles, and even in preserved specimens they are
only to be found by careful examination. What their appearance and
form may be in the living condition cannot therefore be stated here.

The occurrence of this species in the Bahamas as well as in the
Bermudas suggests the possibility of its identity with one of the forms
described by Duchassaing and Michelotti, and in studying their de-
scriptions two species are found to be worthy of remark in this connec-
tion. The first of these is the Actinia aster mentioned by Ellis ('86)
and described by him as having " a thick, fleshy, smooth, and almost
cylindrical stem, ending abruptly at the top, w^hich is provided with
circular rows of tentacles." This description contains nothing that is
particularly distinctive, but nevertheless Duchassaing and Michelotti
('66) have identified with it a form which they describe as about an
inch in height and with the tentacles ringed with white and brown, while
the body and disk are ^^ color es en rougedtre et en bleu tres-clair.^''
Taking this latter description as a guide for the identification of A.
aster, it will be seen that while making the necessary allowances for the
uncertainties which are associated with color descriptions, there is a
certain vague approximation of the coloration to that of Diplactis
hermudensis ; nevertheless, the discrepancies are too great to allow of
an identification of the two forms. Dr. Northrop's notes contains no
statements as to variations in color of D. hermudensis, and until it is
found that these exist and that they approximate the coloration
described for A. aster, it seems better to consider the two forms dis-
tinct.

Another form also presents possibilities in this connection, viz.
the Anemonia depressa of Duchassaing and Michelotti ('6o), a form
with a large disk yellowish in color, with tentacles shorter than the
diameter of the disk, and tinted with shades of blue and reddish
{^'leurs nuances sont le bleu et le rougedtre'^). A figure is given of the
disk and tentacles of this form and might answer for those of D.
bermudensis, and if the mention of ^'le bleu^^ were omitted, the color
descriptions would correspond fairly well. Of the two possibilities
I think the identity of Diplactis bermudensis with A nemonia depressa
is the more probable one, but even here the uncertainty is too great
to justify the identification without further evidence, and for the
present it seems wiser to allow the name used here to stand.

110 ACTINIANS FROM THE BAHAMA ISLANDS

Family RHODACTID^
Ricordea florida Duch. and Mich, i860
Synonym: Heteranthus floridus (D. & M.) McMurrich. 1889.

In my paper on the Bahama Actiniaria I described this form as a
species of Klunzinger's genus Heteranthus, disregarding the name
bestowed upon it by Duchassaing and Michelotti on the ground that
the characters assigned to the genus by those authors were specific
rather than generic, and therefore insufficient, while Klunzinger's
definition of his genus was quite adequate. Although this criticism
still seems to me to be just, yet, nevertheless, according to the strict
laws of priority, Duchassaing and Michelotti's name is the one which
should be employed, and I take this opportunity of correcting my error
in the matter.

Attention may also be called to the probability of the genus Ho-
mactis established by Verrill ('69), being identical with Klunzinger's
Heteranthus, so that it too has a claim prior to Heteranthus, the
latter genus not having been established until 1877. Ricordea is,
however, prior to both, and therefore should supplant them.

Family ZOANTHID^

Zoanthus njmiphaeus (Les.)

Synonyms: Mammillifera nymphcBa Lesueur. 181 7.
Palythoa nymphosa Dana. 1849.
Palythoa (Mammothoa) nymphosa Andres. 1883.

In 181 7 Lesueur established the genus Mammillifera for the recep-
tion of two Zoanthids (which he named M. auricula and M. nymphcea),
and characterized the genus as containing those forms which possess
" a large cuticular expansion, serving as the base of numerous animals,
which, when contracted, assume the form of mammae." Andres (^S^)
unites this genus with Palythoa, making it a subgenus for which he
proposes the name Mammothoa, but this is evidently an error, since
he characterized the genus Palythoa as having sandy incrustations in
the column walls, while Lesueur distinctly gives it to be understood that
his species of Mammillifera have fleshy walls. Erdmann in '85
revived Lesueur's genus, though, as Haddon and Shackleton (^91)
have pointed out, it is questionable if the form he referred to the genus

ACTINIANS FROM THE BAHAMA ISLANDS 111

can be associated with Lesueur's type species, and it is interesting
accordingly to have an account of the structure of one of Lesueur's
species, in order that the true position of his genus may be determined.

Among the slides which Dr. Northrop had prepared I find a
number of a Zoanthid which he had provisionally designated No. 3,
and also a number of drawings of the same form, one of which was the
figure of the group of individuals taken from preserved specimens
(fig. 4). Unfortunately, in the material forwarded me there were no
examples of this No. 3, but there were specimens of a form which the
accompanying label stated to have been collected by Dr. E. A. An-
drews at Green Turtle Cay, Bahama Islands. This form resembled in
general appearance the drawing of No. 3, and preparations which I
made of it demonstrated with certainty its identity with Dr. Northrop's
No. 3.

As regards its identity with Lesueur's M. nymphcea, there must
necessarily be a certain amount of uncertainty. It agrees with the
figure of M. auricula given in Lesueur's paper, and it answers the
generic description ; unfortunately, I find no memoranda of its colora-
tion, and base the identification with nymphcea, rather than with auri-
cula, on the number of tentacles, which is about fifty, and which
Lesueur states to be about fifty in the former species and from twenty-
six to thirty in the latter.

The individual polyps composing a colony are seated close together
upon a coenenchymatous expansion, and reach in preserved specimens
a height of about 2-3.5 mm., the measurement being taken from the
point of attachment to the coenenchyme. The diameter of the column
is about 3 or 4 mm. at the top, slightly less lower down, and the capitu-
lum shows clearly a number of radiating ridges.

The column wall is smooth and without embedded particles of
foreign matter. In structure it resembles closely what has been
described ior Zoanthus sociatus, the same large lacunar spaces occurring
in the mesogloea, while the ectoderm is enclosed within the outermost
portion of the mesogloea, being covered by a mesogloeal subcuticula,
and by a cuticle, much more distinct in some specimens than in others.
The sphincter muscle, which, for diagnostic purposes seems to be of
great importance in the Zoantheae, is double, the two parts being well
marked off from one another. The arrangement is shown in fig. 5,
and from this it will be seen that the upper portion of the sphincter is

112 ACTINIANS FROM THE BAHAMA ISLANDS

small, while the lower is of some length and in its upper part is com-
posed of a few large cavities which occupy the entire thickness of the
mesogloea, lower down the cavities becoming smaller, but being for
the most part more or less circular. The figure will, however, give a
better idea of its arrangement than many lines of description, and from
it it will be seen that in the number and arrangement of the muscle
cavities the sphincter differs from that of any other form which has
hitherto been described, particularly from that of the form described
by Hertwig ('88), as [Z. Dance j the sphincter of which has been de-
scribed by Erdmann ('85), and which might possibly be considered
identical with the form under consideration.

The tentacles are completely infolded in the majority of the indi-
viduals of the colonies I have examined, being visible, however, in a
few cases. They are short, as is usual in Zoanthids, and are arranged
in two cycles, their number being in the neighborhood of fifty — fifty-six
in one specimen in which they could be counted. Nothing worthy of
note was observed in connection with the disk, but in connection with
the stomatodaeum it may be stated that its siphonoglyph was deep,
about one-third of the total surface of the stomatodaeum being occu-
pied by its smooth walls.

The mesenteries are arranged in the microtype (brachycnemic,
Haddon), and there are about twenty-eight to thirty- two pairs. Each
has a single lacunar space near its point of attachment to the column
wall, and its mesogloea is thinner than is usual. Three of the individuals
were sexually mature, and it is interesting to note that both ova and
spermatozoa were present in the same individual, this species, like that
observed by Erdmann ('85), being hermaphrodite.

Isatirus Duchassaingi (Andres)

Synonyms: Zoanthus tuberculatus Duch. & Mich. i860.
Antinedia tuberculata Duch, & Mich. 1864.
Antinedia Duchassaingi Andres. 1883.

This species was originally described by Duchassaing, in a paper
('50) to which I have notjaccess at present, as Zoanthus tuberculatus,
the generic name being supplanted in a subsequent paper ('64) by
Antinedia. In my paper on the Actiniaria of the Bermudas ('89a)
I considered this latter genus to be identical with the genus Mammilli-
fera as defined by Erdmann ('85) ; since then Haddon, in a paper writ-

ACTINIANS FROM THE BAHAMA ISLANDS 113

ten in conjunction with Miss Shackleton ('91), has shown that there
is little probability that the species assigned to the genus Mammillifera
by Erdmaun present the characters of the forms upon which that genus
was founded by Lesueur ('17), and employs the generic term used by
Gray ('28) for a tuberculate Zoanthid. I agree with Haddon's con-
clusions as to this matter, and have therefore followed his example in
using the name Isaurus.

The form described by Gray was from an unknown locality and was
given the specific name tuherculatus. I found among the Bermudan
forms which I studied a species which seemed to agree closely with
Gray's descriptions, and I identified it with his species. At the same
time I also identified the species described by Duchassaing and
Michelotti with this same form, an identification I now find to have
been erroneous. The tuberculate Zoanthid in the Northrop collection
is undoubtedly identical with the form described by Duchassaing and
Michelotti, but it presents unmistakable differences from the Bermuda
species, and must be regarded as distinct from it. The specific name
given by Duchassaing and Michelotti being preoccupied by Gray's
species, a new name must be bestowed, and the term proposed by
Andres ('83) seems most appropriate. Duchassaing and Michelotti's
species may therefore be known as Isaurus duchassaingi.

The appearance of the living animal in its contracted condition is
shown in fig. 6, and I take the following description of it from Dr.
Northrop's notes. The animals are found growing in groups of five or
six, for the most part disconnected, though one specimen had con-
nected with it a smaller individual, evidently produced by budding,
and it may be presumed that each group owes its existence to this
process, the various individuals separating from one another after a
time. Duchassaing and Michelotti figure two individuals united by a
stolon-like coenenchyme, but in the preserved individuals I examined
separation had taken place. In color the polyps were yellowish, closely
resembling their surroundings, the disk and tentacles being of the same
color as the column. The base is firmly adherent, and the column is
"nearly cylindrical, but generally contracted near the base and often
at intervals above, giving the animal a rather grotesque appearance.
When slightly contracted the column bears a number of tubercles,
which, though irregularly distributed, are more numerous near the
top. In preserved and contracted specimens these tubercles are

114 ACTINIANS FROM THE BAHAMA ISLANDS

quite conspicuous." I can add to this statement as to the distribution
of the tubercles the fact that the tubercles are more numerous and more
perfectly developed upon one side of the column than on the other, the
arrangement described by Haddon and Miss Shackleton ('91) for
I. asymmetricus being thus recalled.

The column wall is very much thinner than that of /. tuherculatus,
and in addition does not show any of the large canals filled with cells, so
abundant in the Bermudan form. Numerous canaliculi and scattered
cells, the latter frequently placed in the canaliculi, occur, the mesogloea
being otherwise structureless. The ectoderm of the column presents
the arrangement which is usual in Zoanthids, being covered by a layer
of mesogloea, the subcuticula, from which trabeculae extend inwards
to unite with the general mesogloea, cavities being thus formed which
inclose packets of ectoderm cells (fig. 7). Externally to the subcuticula
is a delicate cuticula to which foreign particles adhere, and the ectoderm
cells contain numerous Zooxanthellae. This arrangement is identical
with what I have described for /. tuherculatus ('89a), and agrees in its
essential particulars with the accounts of the majority of the authors
who have given the matter their attention. Quite recently Von Heider
('95), in his account of Zoanthus chierchice, has come to the conclusion
that the trabeculae which separate the various packets of ectoderm are
cells, and that the subcuticula is formed by the fusion of their outer
ends. Such an interpretation of the appearances is certainly not
applicable to either of the species of Isaurus I have examined; the
trabeculae are so thick that their continuity with the mesogloea is be-
yond question, and, furthermore, the structure of the subcuticula shows
it to be identical with the matrix of the mesogloea. Nor do the
appearances presented by such a form as Zoanthus sociatus warrant
the belief that the trabeculae are otherwise constituted than in Isaurus,
and so far as my observations are concerned they tend to show that
the conclusions of Von Heider are erroneous.

The tubercles are thickenings of the mesogloea, but each contains
also a cavity lined with cells continuous with the endoderm. Dr.
Northrop's notes contain the following description of the tubercles:
" In places the mesogloea contains on its inner side deep cavities which
axe lined by the endodermis. The muscular layer of the mesogloea,
however, does not enter the cavity, but forms a ring aroimd its base
(i.e. its mouth), so that when contraction takes place, the base of the

ACTINIANS FROM THE BAHAMA ISLANDS 115

cavity is drawn up like a purse and the body wall bulges out and forms
a tubercle."

The upper portion of the column, which, in completely contracted
individuals, forms the apex, is destitute of tubercles and is furnished
with numerous longitudinal ridges. The sphincter muscle is strong
and has the form represented in fig. 8, being simple, though showing
a tendency to be divided into two parts, an evident constriction (cr)
occurring upon it. This is not an individual peculiarity, as I have
found it in three individuals examined, and it is interesting as forming
an intermediate condition between the simple sphincter of I. tuhercula-
tus and the double one of the genus Zoanthus.

The tentacles are arranged in two rows, and their number seems to
be double that of the pairs of mesenteries.

The mesenteries are arranged on the microtype, and vary in num-
ber in the different individuals from eighteen to twenty-three pairs.
They are much more slender than those of /. tuberculatus, and though
thickened towards their insertion with the column wall, yet this thick-
ening is slight when compared with that of tuberculatus. Two or three
slightly elongated (in section) canals are found in the thickened portion
of the perfect mesenteries, the imperfect ones containing but one canal.
No reproductive organs occurred in any of the specimens examined.

The stomatodasum presents a well-developed siphonoglyphe, and
its surface is thrown into numerous marked ridges which correspond
in a general way, but not accurately, with the intervals between suc-
cessive pairs of perfect mesenteries.

LIST OF REFERENCES

1886. Ellis, J. — The Natural History of Many Curious and Uncommon Zo-
ophytes collected from Various Parts of the Globe, systematically arranged
and described by Daniel Solander. London. 1786.

1817. Lesueur, C. a. — Observations on Several Species of the Genus Actinia;
illustrated by figures. — Journ. Acad. Nat. Sci. Philadelphia. Vol. I.
1817.

1828. Gray, J, E. — Spicilegia Zoologica. London. 1828.

1850. DucHASSAiNG, P. — Animaux radiaires des Antilles. Paris. 1850.

i860. DucHASSAiNG, P. ET MiCHELOTTi, G. — Memoire sur les Coralliaires des
Antilles. — Mem. de VAcad. Sci. Turin. Ser. II. Tome XIX. i860.

1864. DUCHASSAING, P. ET MiCHELOTTi, G. — Supplement an Memoire sur les
Coralliaires des Antilles. Mem. de VAcad. Sci. Turin. Ser. II. Tome
XXII. 1864.

116 ACTINTANS FROM THE BAHAMA ISLANDS

1882. Hertwig, R. — Report on the Actiniaria. Scient. Results of the Voyage of
H. M. S. Challenger during the Years 1873-1876. Zoology. Vol. VI,
Part XV. 1882.

1883. Andres, A. — Le Attinie. Atti dei Lincei. Ser. 3. Vol. XIV. 1883.
1885. Erdmann, a. — Ueber einige neue Zoantheen. Ein Beitrag zur ana-

tomischen und systematischen Kenntniss der Actinien. Jenaische Zeit-
schr. Naturwiss. Bd. XIX. 1885.

1888. Hertwig, R. — Report on the Actiniaria. Supplement. Scient. Results
of the Voyage of H. M. S. Challenger during Years 1873-1876. Zoology.
Vol. XXVI, Part LXXXIII. 1888.

1889. McMuRRiCH, J. P. — The Actiniaria of the Bahama Islands, W.I. Jour-
nal of Morphology. Vol. III. 1889.

1889a. McMuRRiCH, J. P. — A Contribution to the Actinology of the Bermudas.
Proc. of the Acad. Nat. Sci. Philadelphia. 1889.

1890. Wilson, H. V. — On a new Actinia, Hoplophoria coralligens. Studies
from the Biolog. Labor. Johns Hopkins Univ. Vol. IV. 1890.

1891. Haddon, a. C, and Shackleton, Alice M. — Reports on the Zoological
Collections made in Torres Straits, by Professor A. C. Haddon, 1888-1889.
Actiniae. I. Zoantheae. Scient. Trans. Royal Dublin Soc. Ser. II. Vol.
IV. 1891.

1893. McMuRRiCH, J. p. — Report on the Actiniae collected by the U. S. Fish
Commission Steamer Albatross^ during the Winter of 1887-1888. Proc.
U. S. Nat. Museum. Vol. XVI. 1893.

1895. Von Heider, A. — Zoanthus Chierchiae, n. sp. Zeitschr. fur wiss.
Zoologie. Bd. LIX. 1895.

EXPLANATION OF PLATE XVII

Fig. I. Transverse section through tentacle of Heteractis lucida. /6 = tubercle.
Zeiss C 2.

Fig. 2. Transverse section through a pair of mesenteries of the first cycle of
Heteractis lucida. Zeiss A 2.

Fig. 3. Diplactis bermudensis, natural size. Drawn by Mrs. Northrop.

Fig. 4. Zoanthus nymphcBus, x 3. Drawn by Mrs. Northrop.

Fig. 5. Longitudinal section through the upper part of the column wall of Zoan-
thus nymphceus, showing the sphincter. sp' = upper and sp = lower part
of the double sphincter. Zeiss a 2.

Fig. 6. Isaurus Duchassaingi, natural size. Drawn by Mrs. Northrop.

Fig. 7. Portion of a transverse section through the column wall of Isaurus Du-
chassaingi. cw = cuticula; jcw = subcuticula ; ec = ectoderm; m^ = meso-
gloea.

Fig. 8. Longitudinal section of the upper part of the column wall of Isaurus
Duchassaingi, showing the sphincter, cr = constriction almost dividing
the sphincter.

ACTINIANS FROM THE BAHAMA ISLANDS 117

APPENDIX

Since the above notes were published in 1896 additional observations have
shown the advisability of modification of the systematic position assigned to two
of the forms described.

Heieractis lucida was retained in the genus to which Duchassaing and Miche-
lotti had referred it thirty years earlier, on account of the peculiar structure of the
tentacles. Duerden/ however, in 1898, called attention to the fact that the tubercles
borne upon the tentacles were equivalent to the bands of nematocysts which occur
upon the tentacles of another West Indian form, Aiptasia annulata Lesueur,
and consequently assigned it to the same genus, repeating his conclusion that this
was its proper position in a subsequent paper.^ In its other structural peculiari-
ties, notably in its weak mesoglceal sphincter, lucida closely resembles other species
of Aiptasia, and I now agree with Duerden that it ought to be assigned to that
genus.

The other species which calls for mention is Diplactis hermudensis. My
original description of this form was based upon specimens from the Bermudas,
and in 1898 Verrill ' described a form from the same locality which is apparently
identical, so far as may be judged from the description of the external form and
coloration which is all that Verrill's account includes. Verrill, however, assigns
his form to the genus Actinia, naming it A. hermudensis, and this, I now believe,
is the position to which my Diplactis hermudensis should also be assigned, its
structural peculiarities being more in harmony with those of the genus Actinia
than with those of the family Phyllactidae to which I originally assigned it. I have
already stated elsewhere my altered opinions on this matter and merely reaffirm
them here.^ The correct name for the species is Actinia hermudensis (McMurrich)
Verrill, and it should therefore be referred to the family Actinidae instead of to the
Phyllactidae.

* Duerden, J. E., "The Actinaria around Jamaica," Journ. Inst, of Jamaica, II,
1898, p. 457.

^ Duerden, J. E., Report on the Actinians of Porto Rico, Bull. U. S. Fish Com. for
1900, Vol. 2, p. 355. Washington, 1902.

^ Verrill, A. E., Descriptions of new American Actinians, with critical notes on
other species I. Am. Journ. Sci. (IV), VI, 1898, p. 495.

* McMurrich, J. P., " A Revision of the Duchassaing and Michelotti Actinian types
in the Museum of Natural History, Turin," Boll. Mus. Zool. ed. Anat. Comp. R.
Univ. di Torino, XX. 1905.

118

FLORA OF NEW PROVIDENCE AND ANDROS ^

WITH AN ENUMERATION OF THE PLANTS COLLECTED BY
JOHN I. NORTH^ROP AND ALICE R. NORTHROP IN 1890

By Alice R. Northrop

Introduction

During the latter part of 1889 a report of the work in the Bahamas
of the Danish botanist, Baron Eggers, was received at the herbarium
of Columbia University, and with it a letter from Sir William Thiselton-
Dyer, expressing a hope that American botanists would continue the
exploration. My husband, Dr. John I. Northrop, Instructor in Zo-
ology at Columbia University, was at that time contemplating a
Southern trip for the purpose of studying and collecting marine inver-
tebrates. Sir William Thiselton-Dyer's letter was brought to his notice,
and the result was that a Bahaman trip was planned with both objects
in view. Over six months were spent on the islands, from January to
July, 1890. Of this time two months were passed on the island of
New Providence, where the time was mainly taken up with zoological
work, and the remainder on Andros, where the most interesting plant
collections were made.

In order to understand properly the distribution of the plants and
the relations of the flora, it will be necessary to give a general idea of
the position and conformation of the two islands visited. New
Providence is one of the smaller islands of the group, being only about
twenty miles long and seven wide. It lies on the northern edge of a
portion of the Great Bahama Bank. Nassau, the seat of government
and a well-known health resort, is situated on the slope of a ridge that
runs along the northern shore of the island. The highest point of this
ridge. Fort Fincastle, is about one hundred feet above sea-level. From
here one has an excellent view of the city and harbor, the latter pro-
tected by the narrow outlying cays known as Hog Island, Long Island,

^ Memoirs Torrey Botanical Club, Vol. XII, No. i, 1902.
119

120 FLORA OF NEW PROVIDENCE AND ANDROS

or Quarantine, and, farther seaward. Salt Cay. In the opposite direc-
tion a low level country, covered with trees and dotted here and there
with cocoanut groves, stretches away to the Blue Hills.

Roughly speaking, the physical features of New Providence may be
described as a rocky ridge, about one hundred feet above sea-level at
its highest part, extending along the north side and covered with a
growth of angiospermous trees and shrubs ; a low central plain out of
which rises a second ridge, the Blue Hills, like the first, but narrower
and lower ; then a slightly undulating region covered with the Bahama
pine, extending to the low and swampy south shore. The depressions
of the central plain are occupied by two quite large bodies of brackish
water. Lake Cunningham and Lake Killarney. The latter is the larger
and contains numerous mangrove islets.

The rock of both islands is of ^Eolian formation ; it is very hard at
the surface, but becomes so much softer below that it is sawn into
blocks for a building stone. The surface erosion is most striking and
characteristic. In many places the rocks are fairly honeycombed with
holes, pits, and cavities of all sizes; often sharp jagged points project,
making walking extremely difficult. The largest of the pits are locally
known as "banana holes" because they usually contain considerable
earth in which the people plant their bananas. They vary greatly in
size and shape, the majority being probably from eight to ten feet in
diameter ; they are occasionally twenty feet in depth, but are usually
much shallower. Their sides are often lined with delicate ferns, many
of which grow nowhere else.

There is little or no soil. Mark Catesby, the first naturalist to
visit the islands, wrote in 1754: ''The Bahama Islands may not only
be said to be rocky, but are in reality entire Rocks, having the surface
in some Places thinly covered with a light Mould which in a series of
Time has been reduced to that Consistence from rotten Trees and
other Vegetables. Thus much of the Character of these Islands being
considered, one would expect that they afforded the disagreeable Pros-
pect of bear Rocks : But on the Contrary they are always covered with
perpetual Verdure and the Trees and Shrubs grow as close and are as
thick cloathed with Leaves as in the most luxuriant Soil." In some
places the soil is reddish, and this is considered the most fertile.

Six weeks of our stay were spent at Ryswick, a country place that
we rented on the shore about three miles east of Nassau. Although

FLORA OF NEW PROVIDENCE AND ANDROS 121

the greater part of this time was given to the zoological work, still
between times we collected over two hundred species of plants, cross-
ing the island several times and exploring it in many directions. The
collection, of course, included many cosmopolitan weeds and intro-
duced plants that were found in Nassau and its environs. Among the
latter, growing commonly about the city, were the glossy-leaved almond
tree (Terminalia Catappa), the graceful Spanish cedar (Casuarina
equisetifolia) , the buttressed ceiba or silk-cotton tree, the sand-box tree
{Hura crepitans), and the beautiful flamboyant {Poinciana regia)
with its fern-like foliage.

Having completed the zoological work that had been planned, we
made a diligent study of the chart, and finally decided to visit Andros
next as the largest and least known of the islands, and the one from
which no botanical collections had ever been made. Although the
nearest part of Andros is only twenty -five miles from New Providence,
we could get but little information concerning it until we met Mr. Alex-
ander Keith, of Edinburgh, who had a sisal plantation on Andros.
To him we were indebted for many favors both at this time and later.
A "norther" delayed our sailing for ten days, but we finally reached
Andros March 14, and remained there until July 3.

Andros is by far the largest island of the group, being nearly one
hundred miles long and forty or fifty wide in its broadest part and hav-
ing an area of over nineteen hundred square miles. It is in reality not
one island but a group of islands, the larger northern portion being
separated from the southern and central parts by shallow channels
known as "bights." There is a northern, a middle, and a southern
bight, but they are so filled with cays that the whole archipelago, as it
might be termed, is called by the general name of Andros. In its
prominent physical features Andros resembles New Providence,
although its greatest length runs north and south instead of east
and west, as in the case of New Providence. It has a rocky ridge
extending along the east coast, except at the extreme southern end,
extensive pine barrens in the interior and low mangrove flats on the
opposite side. On Andros, however, the last cover a much greater
portion of the island and constitute its most characteristic feature.
The local name of this region is "swash," ^ a most appropriate term, as
in wet seasons it is half under water. The pine belt is always spoken

^ See illustration — page 218.

122 FLORA OF NEW PROVIDENCE AND ANDROS

of as the "pine-yard,"^ and the hardwood growth on the rocky,
elevated portion is called the "coppet."

The pines are most abundant on the northern part of the island,
and at the extreme southern end, below Grassy Creek, where the rocky
ridge is wanting, there were said to be no pines. None could be seen
from the shore, but we did not cross the island so far south. At Nicol's
Town, the most northerly settlement, the belt of coppet is only about
three quarters of a mile wide ; the pine-yard then begins and extends
to the swash on the other side. At Conch Sound, a few miles south of
Nicol's Town, the pines come down to the eastern shore, but below
Mastic Point, the next settlement, the belt of coppet becomes much
wider. The swash is more extensive than the other two regions put
together and covers hundreds of square miles ; next in extent are the
pine barrens, while the coppet is scarcely more than a comparatively
narrow belt or fringe along the east coast. The pines sometimes extend
in long points far out into the swash.

Numerous creeks drain the island, the majority being on the east
side ; in very wet seasons there is said to be water communication be-
tween those of the east and west side. The creeks are generally nar-
row and winding, but they occasionally spread out into lakelike expan-
sions in the interior. The largest of these lakes are on the west side,
near Wide Opening. A number of the creeks are fresh at their source.

All the settlements are on the east side, with the exception of a small
one at Red Bays on the northwestern end of the island. Nicol's Town
is one of the largest, and when we were there had about three hundred
inhabitants. At the time we visited Andros there were but seven white
people on the entire island. There were no roads, and communication
between the settlements was entirely by water, the reef making a safe
channel for small boats all along the eastern coast. The west coast is
exceedingly shallow, so much so that our boat, drawing only about two
feet of water, had sometimes to anchor a quarter of a mile from the
shore. Even at the edge of the Great Bahama Bank, sixty or seventy
miles farther west, there are but three or four fathoms of water. The
only visitors to this coast are the "spongers."

During the four months spent on Andros we explored it quite
thoroughly, crossing it several times and almost circumnavigating it,
making stops at the various settlements on the way or camping out on

* See illustration — page 219.

Alain Street in Xicol's Town.

FLORA OF NEW PROVIDENCE AND ANDROS 123

the west side where there were no settlements. From one to six weeks
were spent at each of the following places: Nicol's Town, Conch
Sound, Mastic Point, Fresh Creek, Lisbon Creek, and Deep Creek
on the east side, and Red Bays on the west. We sailed through the
northern and middle bights and partly through the southern, and pene-
trated the following creeks, most of them to the head of navigation for
a rowboat ; London, Stafford, Fresh, Lisbon, Deep, and Grassy creeks
on the east ; Loggerhead and Big Cabbage creeks and Wide Opening
on the west.

Botanical Regions

The following botanical regions, each with markedly characteristic
plants, were well defined on both islands : First, the maritime or coast
flora of the northern side of New Providence and the east side of
Andros. These shores were rocky with scattered sandy beaches. The
following plants were common on both islands: the sea-grape
{Coccolohis uvifera), the buttonwood (Conocarpus erecta), the sandfly
bush (Rhacciallis rupestris), and Strumfia maritima. The wild sapo-
dilla (Mimusops dissecta), Joe-bush (Jacquinia Keyensis), Cordia
Sebestena, Borrichia arborescens, and the ram's horn {Pithecolobium
Unguis-cati) were always found near the shore, while on the sandy
beaches flourished the cocoa-plum {Chrysobalanus Icaco), SccBvola
Plumieri, Suriana maritima, Tournefortia gnaphaliodes, the bay
lavender {Ambrosia hispida), Euphorbia buxifolia, the widely dis-
tributed Salicornia ambigua, Sesuvium Portulacastrum, Cakile CBqua-
lis, and the horse-bean (Canavalia obtusifolia) .

Second, the ''coppet," or growth of angiospermous trees and
shrubs found on the more elevated parts of the islands and on the
rocky ridges. The highest elevation on either island was about one
hundred feet, but the ridge was in most places considerably lower than
this. On Andros the highest point was near the center of the island,
at the mouth of the northern bight, marked as Salvador Point on the
charts, but locally known as Bearing Point. The surface erosion was
much more marked on Andros than on New Providence. In some
places the rock was honeycombed with pits of all sizes, in others it
was covered with sharp knife-like projections. Banana holes were
far more numerous. On both islands the elevated ridges, covered with
the *'coppet," showed the greatest amount of erosion. The trees

124 FLORA OF NEW PROVIDENCE AND ANDROS

most commonly met with were the gum elemi {Bur sera Simaruha)^
the poison wood {Metopium), the wild cassada {Dipholis salicifolia),
the horseflesh (Lysiloma paucifoliola) , and the Madeira (Swietenia
Mahogani). As a rule the trees were comparatively small, not more
than a few inches in diameter. The largest and tallest were seen in
what was called the "high coppet" near Deep Creek, Andros. One
horseflesh there measured five and a half feet in circumference at a
distance of four feet from the ground, another six feet four inches,
while the largest mahogany seen was between two and three feet in
diameter. Common among the underbrush were the cockspur thorn
{Pisonia aculeata), the chawstick (Gouania Domingensis), hardhead
{Phyllanthus epiphyllanthus), Erithalis fruticosa, and Duranta re pens.
Among the climbing plants the dream vine {Eckites- umhellata)^
Triopteris rigida, and Ipomxa sinuata were common. The coppet
was usually quite difficult to penetrate, the trees being mostly small
and close together and the underbrush dense.

The third region was the " pine-yard " or pine barrens. This was a
comparatively level region occupying the interior of both islands and
covered almost exclusively with the Bahaman pine {Pinus Bahamen-
sis). Where the ground was a little elevated there were small coppets
or islands, as it were, of angiospermous trees ; where it was lower and
more moist, occasional clumps of palmetto varied the monotony. The
Bahaman pines are tall and slender and do not branch until quite near
the top. The tallest we saw was about seventy or eighty feet in height,
and the largest was four feet and nine inches in circumference. They
do not grow close together, but are usually from ten to twenty feet
apart even when small. A tall brake known as the "May-pole"
{Pieridium caudatum) was very characteristic of the pine belt. It
often formed almost impenetrable thickets six or seven feet in height,
while at one place on Andros we found it growing nine feet in height.
The cinecord {Acacia choriophylla) was common in the pines, as
were also, among the lower plants, Ascyrum hypericoides, Tetrazygia
bicolor, Linum Bahamense, Ernodea littoralis, and Vernonia Bahamen-
sis. The showy sedge {Dichromena colorata) and the purple orchid
{Bletia verecunda) were abundant in the pines and were also occasion-
ally found on the savannas. In many parts of the pine barrens on
Andros there was no underbrush, nothing but a coarse grass called
"bed-grass" (a species of Andropogon), relieved here and there by

Tn the "High Coppet" near Deep Creek.

FLORA OF NEW PROVIDENCE AND ANDROS 125

the crimson flowers of Ipomosa repanda. As one approached the
western edge of the pines, the ground became less rocky, the trees
smaller and smaller, and the palmettoes more numerous until one
finally emerged on either swash or savannas.

The savannas, constituting a fourth distinct botanical region, were
found only on Andros. They were level prairie-like stretches, lying, as
a rule, between the pines and the swash. They were most common in
the northwestern part of the island. The ground was not rocky and
was covered, for the most part, with a coarse sedge called ^'saw-grass"
(Cladium Jamaicense) ; there were also occasional clumps of palmetto
or brier tree (Terminalia spinosa). This region proved excellent
botanizing ground, and by far the greater number of the plants found
there were met with nowhere else. Flaveria linearis, Polygala Boy-
kinii, Eustoma exallafum, Aletris bracteata, Gyrostachys tortilis, and
Gerardia purpurea were common and in some places Limodorum
tuberosum, Buchnera elongata, and Samolus ebracteatus.

The fifth plant region was the "swash." On Andros this region,
as has been said, was very extensive and comprised hundreds of square
miles. Here the eroded coral rock, such a prominent feature of the
coppet and the pine barrens, was replaced by soft, calcareous mud, in
some places more or less hardened, in others very soft. There were
numerous ponds and lakes in this region which we were told became
connected in wet seasons, making a network of waterways navigable
by small boats for many miles. We were there in a comparatively dry
season, and the ponds were very shallow, having about three inches of
water and eighteen inches of marl.

The scenery was monotonous and desolate. In many places, as
far as the eye could reach, the ground seemed perfectly flat and covered
with small mangroves, the salt-bush {Avicennia nitida), and a low
form of buttonwood (Conocarpus erecta), none more than a few feet
in height. The plants were in reality quite scattered and a consider-
able distance apart, but seen at a distance, the effect was that of a
smooth expanse of lawn. Here and there a dark line of pines showed
on the horizon or one caught the gleam of water, but as a rule only
clumps of palmettoes or a few shrubs varied the monotony. In some
places, especially near the creeks, palmettoes were abundant, the
most common being the silver thatch {Thrinax Bahamensis), the
hog cabbage {Cyclospathe Northropi), and the saw-tooth cabbage

126 FLORA OF NEW PROVIDENCE AND ANDROS

{Paurotis Androsana) were occasionally seen; all were of small size.
Toward the southern end of the island the mangroves sometimes at-
tained considerable size and formed the most prominent feature of
the landscape. This desolate, uninhabited region is a paradise for
water birds, which were found here in great numbers. The flamingoes
were the most interesting, and these we often saw while on the west
side pf the island.

I have described the botanical regions in such detail because since
we collected in the Bahamas many of the localities we visited have been
destroyed, botanically speaking, by being cleared for the cultivation of
sisal. The work was just beginning when we were there, a few sisal
plantations having been started on both New Providence and Andros.
Several years later thousands of acres had been cleared and planted
with sisal. For this purpose both coppet and pine barrens were avail-
able, but not the swash. Large companies were formed, a great amount
of money was spent in clearing and planting, in making roads, and I
believe even a small railroad was built on Andros. It was confidently
expected that large fortunes would be made, but after three or four
years' trial these hopes proved to be visionary, and I have since heard
that many of the plantations have been given up and the land allowed
to lapse into its former wild state. It is highly probable, however, that
the flora of Andros has suffered more or less change through the exten-
sive clearing and the probable introduction of cosmopolitan weeds.

Previous Collectors

Mark Catesby explored and collected along the southern Atlantic
coast from 1 731 to 1743 and during that time made a trip to the Baha-
mas, visiting New Providence and also touching incidentally at Andros.
Some of the plants he collected were figured in his ''Natural History of
Carolina," published in 1754. The next record we have of Bahaman
plants were the collections sent to Sir William Hooker by Mr. Swainson
between 1838 and 1842. These were described by Grisebach and in-
corporated in his "Flora of the British West Indies," published in 1864.
Less than two hundred species were there recorded from the Bahamas.
Between 1880 and 1887 Mr. L. J. K. Brace, of Nassau, sent to Kew
through Governor Robinson a large number of Bahaman plants. A
list of these has been incorporated in a Provisional List of the Plants

FLORA OF NEW PROVIDENCE AND ANDROS 127

of the Bahama Islands, by Gardiner, Brace, and DoUey, which was
published by Dr. Dolley in the Proceedings of the Academy of Natural
Sciences of Philadelphia in 1889. This list, however, is not always
clear as to which are native and which cultivated species, and in the
majority of cases the place of collection is not given. A small collection
of plants made by a Mr. Cooper were presumably sent to Dr. Torrey
at Columbia University, as they form a part of the Torrey herbarium.
With very few exceptions all the above collections were made on the
island of New Providence.

In 1887 a grant was made by the British Association, for the inves-
tigation of the Bahaman flora, and the Danish botanist. Baron Eggers,
undertook the work. He spent from November, 1887, to April, 1888,
in the islands and brought back 314 species. A few were collected on
Fortune Island and Long Island, but the great majority were from New
Providence. Professor T. H. Herrick, of Johns Hopkins, visited Abaco
in 1886 and collected a small number of plants noted in the Johns
Hopkins Univ. Circ. 6, 46. During the winter of 1 890-1 891 Professor
Albert S. Hitchcock, of the Missouri Botanical Garden, accompanied
a party of naturalists, headed by Dr. J. T. Rothrock, of the University
of Pennsylvania, on an exploring trip through the Bahamas. Eleu-
thera. Cat Island, Watling's Island, Crooked and Fortune islands, and
Inagua were visited, as were also the islands of Jamaica and Grand
Cayman. The report of the plants collected was published in 1893 by
Professor Hitchcock in the Report of the Missouri Botanical Garden.
The total number of plants there noted from the Bahamas was 380,
and of these two were described as new.

Analysis of the Collection

The collection enumerated in the following pages consists of 542
species (461 exclusive of the cryptogams) to which are to be added six
varieties and twenty-one cultivated plants. Two of the collection
proved indeterminable on account of insufficient material, while fifteen
could only be determined generically for the same reason. The total
number of families of flowering plants represented is 93, the number of
genera 304. The families most largely represented are Leguminosae,
with 45 species ; Compositae, with 34; Rubiaceae, with 24, and Euphor-
biaceae, with 21; while Orchidaceae, Convolvulaceae, and Verbenaceae

128 FLORA OF NEW PROVIDENCE AND ANDROS

come next. The genus most largely represented is Ipomoea, of which
we collected 13 species; eight species of Cassia were found, and six
species each of Euphorbia, Coccolobis, and Tillandsia, while Passi-
flora and Eupatorium each have five species. As will be noticed, there
is a very large proportion of genera to the number of species, in the
majority of cases a genus being represented by but a single species.

Of the plants collected a new Chara was described and published
by Dr. T. F. Allen, an Anastraphia by Mr. J. M. Greenman, of Cam-
bridge, a Jacquinia by Professor Mez, and new species of Ccesalpinia,
Phyllanthus, Reynosia, and Casearia by Professor I. Urban, of Berlin
in " Symbolae Antillanae." In addition new species of Hyrnenocallis ,
Aletris, Vanilla, Phoradendron, Pithecolobium, Cassia, Linum, Ery-
throxylon, Crossopetalum, Rhamnidium, Helicteres, Xylosma, Termi-
nalia, Heliotr opium, Tecoma, Catesbaa, Myrstiphyllum, Anguria,
Metastelma, and Eupatorium are described in this paper, as well as
two new genera of palms, Paurotis and Cyclospathe. The type
specimens are in the herbarium of Columbia University. Sets are also
at Kew, the Royal Botanical Garden at Berlin, the Gray Herbarium,
the Field Museum of Natural History, and Geneva. As far as I can
discover, the following genera have never before been reported from
the Bahamas: Coccothrinax, Inodes, Aletris, Vanilla, Broughtonia,
Polystachya, Cranichis, Limodorum, Hypoxis, Pedilanthus, Maba,
Mitreola, Voyria, Trianosperma, and Aster.

My sincere thanks are due those who have assisted me in the
preparation of this report, especially to the following specialists who
kindly determined the cryptogams : Mr. Frank S. Collins, of Maiden,
Mass., the algae; Dr. Albert Schneider the lichens; Professor Lucien
M. Underwood the fungi ; Mrs. Elizabeth G. Britton the mosses ; and
Professor D. C. Eaton who determined a number of the doubtful ferns
in 1890. The report on the palms has been prepared by Mr. O. F.
Cook, of the Department of Agriculture at Washington, to whom I here
w^ish to express my obligations. Dr. Nathaniel L. Britton and Mr.
George V. Nash kindly named the grasses, and Dr. Britton the sedges.
I am also greatly indebted to Dr. Britton for advice on many points
and for his kindness in comparing and identifying a number of our
plants at Kew in 189 1, also to him and to Dr. John K. Small for revis-
ing the nomenclature in many instances, and to Professor Underwood
for revising the names of the ferns and other kindly assistance. A

FLORA OF NEW PROVIDENCE AND ANDROS 129

number of doubtful specimens were compared by me at Cambridge in
1897, and I take this opportunity of thanking Dr. Benjamin L. Rob-
inson and his assistants for the kindness then shown me and for a
number of determinations they were good enough to make for me
the following year. I also wish to acknowledge gratefully several
determinations made for me by the authorities at Kew in 1897.

The new species have been most successfully drawn by Miss Mary
V. Thayer, of Holbrook, Mass., to whom I wish to express my thanks
for her careful work.^

LIST OF PLANTS

THALLOPHYTA

MARINE ALGiE

Determined by Mr. Frank S. Collins, of Malden, Mass., 1891.

CHLOROPHYCE^

Cladophorace^

CH.ETOMORPHA sp. ? Lake Waterloo, near Nassau, Jan. (209).
Cladophora sp. ? Ft. Montagu, Nassau, Jan. (163).

Caulerpace^

Caulerpa clavifera Ag. Quarantine Cay, Nassau, Jan. (182).
Caulerpa ericifolia Ag. Lake Waterloo, Nassau, Feb. (301).
*Caulerpa plumaris Ag. Lake Waterloo, Nassau, Jan. (208).

CODIACE^

Penicillus capitatus Lamour. Quarantine, Nassau, Jan. (187).
•j-Rhipocephalus phcenix J. Ag. Quarantine Cay, Nassau, Jan.
(190).

tllDOTE A CONGLUTINATA Lamour. Salt Cay, Nassau, Jan. (233).

* The long period that has elapsed since the collection of these plants and the
publication of this report may call for a word of explanation. My husband was work-
ing up the zoological collections and I the plants when his sudden death occurred in
June, 1891. Since then ill-health has year after year prevented any continuous work,
and the frequent lapses of time have made much revision necessary. My husband
greatly assisted me in the beginning of the work, and my sole motive in continuing it
was because of his interest in my doing it.
K

130 FLORA OF NEW PROVIDENCE AND ANDROS

Udotea flabellata Lamour. Quarantine Cay, Nassau (183,
300).

•j-Halimeda tridens Lamour. Salt Cay, Nassau. Quarantine
Cay, Nassau, Jan. (191).

*Halimeda tuna Lamour. Salt Cay, Nassau, Jan. (232).

Valoniace^

*Valonia yEGROPHiLA Ag. ? Lake Waterloo, Nassau, Jan. (210).
•j-DiCTYOSPH^RiA FAVULOSA Decne. Quarantine Cay, Nassau
(188).

MiCRODiCTYON Agardhianum Decne. Salt Cay, Nassau, Feb.

(273)-

Dasycladace^

•f Dasycladus occidentalis Harv. Ft. Montague, Nassau, Jan.
(161).

f AcETABULARiA CRENULATA Lamour. Lake Waterloo, Nassau,
Jan. (211).

Cymopolia barbata Lamour. Silver Cay, Nassau, Jan. (156).

Cymopolia Mexicana J. Ag. Silver Cay, Nassau, Jan. (155).

Blodgettia confervoides Harv. Doubtful position and value.
Quarantine (192).

PH.EOPHYCE.E

• FUCACE^

Cystoseira myrica Kiitz. Nicol's Town, Andros, March (337).
*TuRBiNARiA VULGARIS Ag. Quarantine Cay, Nassau, Jan. (185).
Sargassum filipendula laxa J. Ag. Ft. Montague, Jan. (158).
SARGASSUMsp. Salt Cay, Nassau, Feb. (268-275).

DlCTYOTACE^

DiCTYOTA DiCHOTOMA Lamour. Quarantine, Jan. (177).
*DiCTYOTA FASCiOLA Lamour. Salt Cay, Feb. (274).
*Padina pavonia Gaillon. Silver Cay, Nassau, Feb. (154).
ZoNARiA LOBATA Ag. Salt Cay, N.P., Jan. (259) ; Goat Cay,
Andros, June (632).

FLORA OF NEW PROVIDENCE AND ANDROS 131

RHODOPHYCE^

Helminthocladiace^

LiAGORA Cheyneana Harv. Salt Cay, Nassau, Feb. (272).
LiAGORA ELONGATA Zan. Silver Cay, Nassau, Jan. (153).
fLiAGORA VALiDA Harv. Goat Cay, Andros, June (748).

Ch^tangiace^

*Galaxaura lapidescens L amour. Salt Cay, Nassau, Feb.
(260, 271).

Rhodomelace^

Laurencia paniculata J. Ag. Quarantine, Nassau, Jan. (179).
Some specimens near L. ohtusa Lamour.

*DiGENiA SIMPLEX Ag. Quarantine Cay, Nassau, Jan. (180).

fPoLYSiPHONiA Havanensis Mont. Ft. Montagu, Jan. (160,
181).

fPoLYSiPHONiA Pecten- Veneris Harv. Ft. Montagu, Nassau,
Jan. (159).

PoLYSiPHONiA sp. ? Goat Cay, Andros, June (749).

fDASYA GiBBESii Harv. Ft. Montagu, Nassau, Jan. (162).

Ceramiace^

*Centroceras clavulatum Mont. Quarantine Cay, Nassau,
Feb. (302).

CORALLINACE^

LiTHOTHAMNiON ? Quarantine Cay, Nassau, Feb. (304, 305).
*Jania RUBENS Lamour. Quarantine Cay, Nassau, Jan. (178).

Notes on Distribution. — The species of marine algae marked *
are generally distributed in warm waters. The species marked f are
limited to the West Indian and Florida region. ^'Liagora elongata
and Cystoseira myrica are characteristic Red Sea plants; Liagora
Cheyneana is an Australian species. Liagora elongata and Cymopolia
Mexicana are, I think, new to the West Indian region, although it
is now considered rather doubtful whether the last-named species
is distinct from C. barbata.^^ (F. S, Collins.)

132 FLORA OF NEW PROVIDENCE AND ANDROS

CHARACE^

Chara depauperata T. F. Allen. New species. Bull. Torrey
Club, 21, 267. 1894. Fresh Water Pond, Hog Island, N.P., Feb.

(258).

Chara sp. ? In brackish water. Stafford Creek, Andros, May

(547)-

LICHENES 1

Cladonia gracilis (L.) NyL? (Not mature.) Nicol's Town,
Andros (346).

Cladonia Flcerkeana Fr. Nicol's Town, Andros (347).

Cladonia sp. ? On palmetto. Red Bays, Andros, April (484).

Leptogium pulchellum (Ach.) Nyl. Nicol's Town (444).

Leptogium tremeloides (L.) Fr. Nicol's Tow^n (382).

Ramalina pusilla (Prev.) Tuck. Nassau (755).

With the exception of 484, all are quite common and widely dis-
tributed species.

FUNGI 2

ScHizoPHYLLUM ALNEUM (L.) Schrot. Nicol's Town, March (351).
Clathrella crispa (Turpin) E. Fischer. Andros (798).
DiPLOCYSTis Wrightii B. & C. Andros (777). Also reported
from Inagua by Hitchcock.

Trametes cinnabarina (Jacq.) Fr. Andros (793).
AuRicuLARiA auricula (L.) Schrot. Andros (794).
FoMES igniarius (L.) Fr. /Vndros (797).

BRYOPHYTA '

ToRTULA agraria (Sw.) Hcdw. Nassau and Nicol's Town (34,
341, 168).

Hyophila Barbula (Schwaegr.) Hampe. Nassau (169, 170).
Macromitrium insularum Mitt. Nicol's Town, Andros, March

(342,345)-

Sematophyllum sericifolium Mitt. Nicol's Town (343, 530).

Syrrhopodon flavescens Mueller. Nicol's Town, March (344).

OcTOBLEPHARUM ALBiDUM Hedw. Ncw Providcncc, Nicol's

Town, Feb. (348, 746).

^ Determined by Dr. Albert Schneider at Columbia University.
2 Determined by Professor Lucien M. Underwood.
^ Determined by Mrs. Elizabeth G. Britton at Kew.

FLORA OF NEW PROVIDENCE AND ANDROS 133
PTERIDOPHYTA

SCHIZiEACE^

Ornithopteris adiantifolia (L.) Bernh. Common and variable ;
abundant in the pines. Nassau, Jan.; Nicol's Town, March (12,
15. 83, 300).

*Ornithopteris cicutaria (Kunze) Underw. (Anemia cicutaria
Kunze, Linnsea, 9, 22. 1835.) On rocks, not common. Nassau,
Jan. (165).

POLYPODIACE^

*Dryopteris patens (Sw.) Kuntze. In banana-holes and rocks
along roadsides ; common and variable. Nassau, Jan. ; Nicol's Town
and Red Bays, April (173, 240, 441, 469).

*Dryopteris asplenioides (Baker) Kuntze. Clefts of rocks,
uncommon. Indusium small, pinnae narrow and distant. Conch
Sound, March (416).

Tectaria TRiFOLiATA (L.) Cav. On sides of banana-holcs ; not
common. Conch Sound, May (562).

Davallia clavata Sw. Common. Nassau, Jan. (133). Same
as Wright 961.

*Asplenium: dentatum L. Pinnae large and close together. Caves
near Nassau, Feb. (286).

Blechnum serrulatum Rich. Caves near Nassau, Feb. (285).
Same as Fendler 133.

Adiantum tenerum Sw. In banana-holes. Near Nassau, Feb. ;
Red Bays, April (288, 489).

Pteridium CAUDATUM (L.) Maxon. "May-pole." Very common
on both New Providence and Andros ; makes dense thickets in the
pine-yard six to eight feet high (313).

Pteris longifolia L. Common on walls. Nassau, Jan. (18, 94).

ViTTARiA LiNEATA (L.) Sm. On palmcttocs, not common. Red
Bays, April (472).

Cheilogramma LANCEOLATUM (L.) Bloumc. Leaves 10-14 inches
in length. On trees, not common. Nicol's Town, March (356).

AcROSTiCHUM AUREUM L. "Wild ginger." In low ground, not
common. Conch Sound, May; Deep Creek, June (408, 714). 714 was
growing in a banana-hole ; the leaves were shorter than in 408, fertile
almost to the base, and the rachis was deeply sulcate.

134 FLORA OF NEW PROVIDENCE AND ANDROS

Phlebodium aureum (L.) R.Br. Common on palmettoes. Hog
Island, N.P., Feb.; Purser Point, Andros, June (249, 625, 665, 720).
No. 665 is a sport with leaves very glaucous and pinnae deeply crenate.

PoLYPODiuM POLYPODioiDES (L.) A. S. Hitche. On trees, not un-
common. Nicol's Town, April; Conch Sound, May (442).

Campyloneuron Phyllitidis (L.) Presl. In banana holes, not
uncommon. Conch Sound, May (566).

PoLYPODiUM SQUAMATUM L. On trees. Conch Sound, March.
Determined by Professor Eaton (406).

*Phymatodes Swartzii (Baker) Underw. In banana holes, rare.
Leaves difform, there being a number of short and broad sterile leaves.
Conch Sound, May (581). Same as Wright 799.

GoNiOPTERis REPTANS (Sw.) Presl. In banana holes. Conch
Sound, May. Determined by Professor D. C. Eaton (583).

GoNioPTERis REPTANS CORDATA. In banana holes. Conch
Sound, May. Determined by Professor D. C. Eaton (576).

PsiLOTACEiE

PsiLOTUM NUDUM (L.) Griscb. In hollows in trunks of trees, not
common. Cocoanut Point, Andros, April; Fresh Creek, June (515).

SPERMATOPHYTA

CyC ADAGES

Zamia sp. "Bay rush." Common in the pines in certain localities.
Leaves 15-25 cm. in length, leaflets but ten pairs, seldom opposite,
thick, with revolute margins, 3.5-5 cm. long, 6-9 mm. broad. Seems
to be nearest Z. pumila L. ; resembles Wright 3193. Stafford Creek,
Andros, May (550).

CONIFERyE

PiNUS Bahamensis Griseb. Covers large tracts in the interior of
both New Providence and Andros. N.P., Jan. ; Nicol's Town, April;
Conch Sound, May (84, 440).

JuNiPERUS Barbadensis L. Not common. Southwest Beach,
N.P., Feb. ; Nicol's Town, Fresh Creek, June (321,355).

* Species marked with a star were verified by Professor D. C. Eaton, 1891.

FLORA OF NEW PROVIDENCE AND ANDROS 135

TyPHACEiE

Typha Domingensis Pers. Nicol's Town, March (353).

Naiadace^

RuppiA MARiTiMA L. Stafford Creek, Andros, May. Determined
by Rev. Thomas Morong (536).

Gramine^ *

Andropogon sp. "Bed grass." Purser Point, Andros, common
(659). Specimens not in flower.

Paspalum fimbriatum H. B. K. Nassau, Jan. (118).

Panicum divaricatum L. ? (Not in flower.) Deep Creek, Andros,
July (732).

Panicum proliferum Lam. Fresh Creek, June (620).

Panicum sp. Hog Island, Nassau, Feb. (248).

Cenchrus tribuloides L. ''Devil grass." Nassau, Jan.; Deep
Creek, July (148, 719).

Sporobolus Virginicus Kunth. Deep Creek, July. On sandy
shores (728).

Stenotaphrum Americanum Schrank. Common along shore.
Nicol's Town, April (520).

Chloris Swartz;ana Doell. "Finger grass." Nicol's Town,
April (521).

Cha^tochloa glauca (L.) Scribn. Fresh Creek, June (618).

Eragrostis ciliaris (L.) Link. Nassau, Jan. (167).

Uniola paniculata L. "Bay rush." Quarantine Cay, Nassau,
Feb. (314).

Phragmites or Arundo ? Only glumes remaining. Near South-
west Beach, N.P., Feb. (315).

DiSTiCHLis sp. "Rabbit grass." Big Cabbage Creek, June.
Common on edge of swash. Red Bays (485).

Arthrostylidium capillifolium Griseb. ? "Old man's beard."
Not in flower. Nassau; common, climbing over shrubs and bushes.
Branches leafy, leaves fascicled, wiry, filiform (93).

^ Determined by Dr. Nathaniel L. Britton and Mr. George V. Nash.

136 FLORA OF NEW PROVIDENCE AND ANDROS

Cyperace^ ^

Cyperus brunneus Sw. Common. Hog Island, N.P. ; Nicol's
Town, April (435).

Cyperus ferax Rich. Nassau, Feb. (287).

Cyperus ochraceus Vahl. Nassau, Jan. (144).

Eleocharis camptotrichus schweinitzii C. B. Clarke. Conch
Sound, April (745). Determined by C. B. Clarke, Kew, 1891. " Same
as the plant collected in Guadeloupe by Bertero, taken as E. tenuissima
by Boeck., called by me as above; also mixed in Wright 3367 from
Cuba."

Eleocharis capitata (Willd.) R. Br. Freshwater Pond, Hog
Island, Nassau (247).

Eleocharis ochreata Nees. In banana-holes in the pines. New
Providence (327).

DiCHROMENA COLORATA (L.) A. S. Hitch. Common in the pines,
also found in abundance on the savannas at Red Bays, Andros, New
Providence, Jan.; Red Bays, April (100, 466).

FiMBRiSTYLis MONOSTACHYA (L.) Hassk. Fresh Creek, Andros,
June (634).

FiMBRiSTYLis SPADiCEA (L.) Vahl. Mastic Point, Andros, June;
Purser Point, June (596, 667).

Rhynchospora microcarpa Baldw. Red Bays, April (493).
Same as Eggers 4308 from Bahama.

Rhynchospora cyperoides (Sw.) Mart. Nassau, Feb. (288).

Cladium Jamaicense Crantz. "Saw-grass." Fresh Creek, June

(635)-

ScLERiA FiLiFORMis Sw. Mastic Point, Andros, June (603).

Palms from the Bahamas^

The palms have been neglected so generally in botanical collec-
tions that many striking novelties still remain to be secured by those
who brave the inconvenience of handling plants so unmanageable by
ordinary herbarium methods. The present small series of Bahama
palms shows what may be expected in many parts of the tropics, though

^ Determined by Dr. Nathaniel L. Britton.

^ The families Sabalaceae and Arecaceae were contributed by O. F. Cook, Wash-
ington, D.C.

FLORA OF NEW PROVIDENCE AND ANDROS 137

for the benefit of botanists who may wish to emulate the example of
Dr. and Mrs. Northrop it may be permissible to add that when other
material is being secured, ripe fruits, or even the naked seeds, are ex-
tremely desirable, and may usually be picked up at the base of the tree
long after the fruiting season has passed.

The present list recognizes five palms from the Bahamas, though
two of these are not named specifically for lack of adequate material.
One may be the species reported by Professor Hitchcock as Thrinax
argeniea, while our Thrinax Bahaniensis may correspond to his T.
parviflora, though numerous species of this group are doubtless to be
found in the Bahama archipelago. Grisebach reported only Sahal
umhraculifera, a name no longer tenable. It has been stated also that
the Bahamas have a cabbage palm {Euterpe) and a royal palm {Oreo
doxa), but these names are also not available for West Indian palms.
Moreover, it is not known that specimens exist from which better
identifications could be made.

Sabalace^

Thrinax Bahamensis sp. no v.

Leaves and inflorescence resembling Coccothrinax jucunda Sargent
{Bot. Gaz., 27, 89. 1899), but apparently to be associated rather with
Thrinax Keyensis Sargent {Bot. Gaz., 27, 86. 1899) in view of the short
pedicels, distinctly lobed calyx, broad filaments and short styles.

Petiole 48 cm. long, 15 mm. broad at base, narrowed to 12 mm. near
the apex ; equally convex on both sides, becoming flat above toward the
base ; segments of middle of leaf about 53 cm. long, and 32 mm. broad ;
lateral segments reduced to 30 cm. by 5 mm. ; texture thin and brittle ;
venation also closely similar to C. jucunda, but the surface distinctly
less pubescent, or the pubescence much more fugacious, as in other
species of true Thrinax; inflorescence with secondary branches slender,
subtended by narrow scarious bracts, 8 to 10 mm. in length; bracts
with a distinct midvein and a pencil of hairs at the tip; pedicels of
flowers seldom i mm. in length, with 6 distinct subtriangular lobes ;
filaments triangular, often united at base to form a complete cup;
stigma truncate or somewhat funnelform, about 0.5 mm. in length.

This species is evidently much smaller in all its parts than Thrinax
Keyensis. The comparison of its leaves with those of Coccothrinax
jucunda is based on A. H. Curtis's No. 262 from Big Pine Key, which

138 FLORA OF NEW PROVIDENCE AND ANDROS

seems to correspond well with Sargent's description, though there is the
possibility that the leaf and fruit were not taken from the same tree.

Locality. — Big Cabbage Creek, Andros Island, June. Another
specimen (257) from Freshwater Pond, Hog Island, N.P., February,
consists of a leaf and an old inflorescence, the latter with the spathes
still coated in patches with dense white pubescence.

In comparison with Thrinax Ponceana (Bull. Torrey Club, 28, 536.
1 901) from Puerto Rico the leaves of the present species are smaller,
with the petioles less flattened and more distinctly ribbed on the upper
side near the apex. The transverse or oblique veinules are more
numerous and more prominent ; also the veinules of the lower surface,
which lacks the glaucous or waxy covering distinct in T. Ponceana.

COCCOTHRINAX Sp.

A single leaf with the form and veination of C. Garberi (Chapman),
but somewhat less densely pubescent. Locality: New Providence,
Nassau, February, 1890 (No. 284).

Mr. Lyster H. Dewey, of the U. S. Department of Agriculture,
recently brought back from New Providence Island a leaf probably
belonging to a Coccothrinax and popularly called " silver thatch." The
leaves are commonly used for weaving into hats and baskets. The
trunk seldom, if ever, exceeds about 2.5 m., and is about 15 cm. thick.
A photograph secured by Mr. Dewey shows that the surface is largely
free from leaf bases, and fairly smooth, the leaf scars being but slightly
impressed. The diameter seems to be rather uneven, with a tendency
to become somewhat thicker in the middle.

Paurotis gen. nov.

A small, slender palm with spiny petioles like Copernicia, but with
only the primary branches of the slender inflorescence subtended by
spathes.

Paurotis is probably more nearly related to Serenoa than to Coper
nicia, but differs in the larger size, the erect trunk, the stronger ligule,
the absence of the ligule-like inferior scales, the presence of a rudimen-
tary midrib, and in the more deeply divided segments. Inflorescence
much more slender throughout than in Serenoa; flowers much smaller,
with free sepals and short valvate petals.

FLORA OF NEW PROVIDENCE AND ANDROS 139

The long naked and apically scarious and bilabiate spathes are
strikingly different from those of Copernicia. The inflorescence is
much longer and more slender than that of Serenoa, but in other re-
spects has greater resemblance than to that of Copernicia. The leaves,
on the other hand, are more like Copernicia, though the presence of a
true midrib, even if very small, with one or two segments inserted
somewhat above the base may be taken as a further sign of affinity
with Serenoa.

The present genus will probably accommodate the palm from
Puerto Rico {Sintenis, 6512) referred by Professor Drude to Coper-
nicia, but having no spathes on the branches. Grisebach and Wend-
land described from Cuba, Copernicia Wrightii, which may also belong
to Paurotis.

Hitherto Copernicia has been the only known West Indian genus
of fan-palms with spiny petioles. The type of Copernicia is C. cerifera
(Arruda) from Brazil, but the Cuban species of the genus seem to
resemble Paurotis even less than the Brazilian, since they have the
inflorescence more robust and compact and the spathes more strongly
developed. The species listed by the Index Kewensis as Copernicia
maritima (Coryphamaritima H. B. K.) and Copernicia pumos (Corypha
pumos H. B. K.) have smooth petioles according to the original de-
scriptions, and should have been transferred to Thrinax rather than
to Copernicia. Although treated as a synonym of Copernicia in the
Index Kewensis the generic name Crysophila Blume had priority of
publication, as shown by the fact that it is cited by Martius in connec-
tion with the original description of Copernicia. It seems probable,
however, that Crysophila is distinct from Copernicia as indicated by
Drude. Its type, C. nana (H. B . K. ) , came from the region of Acapulco,
Mexico. The petiole is unarmed, and other characters are quite at
variance with those of Paurotis.

Paurotis Androsana sp. nov.

Trunk 3 to 4 m. high, very slender, 5 to 6.5 cm. in diameter, rough
with irregular scalelike leaf bases; leaves tufted, flat, orbicular;
petioles 52 cm. long, 15 mm. thick at base, 10 to 12 mm. at apex, not
including the spines, 5 mm. thick at base, 3 mm. at apex ; upper face
moderately concave, subcarinate in the middle distad; lower face
strongly convex in the middle, concave on each side; upper surface

140 FLORA OF NEW PROVIDENCE AND ANDROS

with fine longitudinal or oblique impressions, doubtless from the next
leaf ; lower surface nearly smooth, very finely grooved longitudinally ;
both surfaces covered with a thin layer of waxy scales ; toward the
margins are small scattering brownish longitudinal scars, more nu-
merous on the upper side ; occasionally there arises from such a scar a
narrow scarious ribbon 2 mm.- or less in length ; these evidently corre-
spond to the peltate scales of the leaf-bases and petioles of Inodes;
margins of petioles thickened, smooth, corneous, in color pinkish
brown (vinaceous-cinnamon, Ridgway) at base, and dark brown distad ;
teeth somewhat irregularly placed, usually about i cm. apart, but
sometimes 2 cm. and sometimes close together or with two points;
points commonly curved forward, sometimes straight or curved back-
ward ; length about 3-5 mm. ; size decreasing toward the apex of the
petiole, but occasional teeth are mere rudiments ; ligule very broadly
subtriangular after losing a rather broad, thin, subscarious margin;
lateral angles of the ligule coalescing with the margins of the lateral
segments as in Thrinax and related genera; each side of this terminal
widening of the petiole bears a strongly curved marginal tooth.

Apex of petiole on the under side, broadly triangular, about 5 mm.
long. The oblique edges which subtend the insertion of the segments
do not, however, meet in the middle; the middle rib is slightly thicker
than the others and has one or two segments inserted on each side
beyond the apex of the triangle, so that Paurotis may be said to have
passed the stage of Thrinax and to have a true midrib.

Leaves about 52 cm. long, composed of about 36 segments; middle
segments longest, the lateral shortened nearly by half. Segments
united for about 20 cm. in the middle of the leaf, but only for i to 3
cm. at the sides. The segments are split 20 cm. or less from the tips.
The margins of the notch are somewhat thickened, but there is no fiber
like that of Inodes. The edges of the ribs both above and below are
nearly square, and bear along the angles rows of irregularly placed
brown scars like those of the margins of the petiole. Both surfaces of
the leaf have a thin film of wax. There are from 8 to 11 longitudinal
veinlets larger than the others and i mm. or more apart ; between these
are similar numbers of finer equal veinlets, the middle one sometimes
slightly larger ; oblique veinlets distinct, but very fine, not close.

Inflorescence 80-90 cm. long, the main axis bearing from 9 to 10
tubular spathes. Base of inflorescence flattened and the lower spathe
with compressed, winglike margins. The spathes are open only at
the bilabiate apex.

FLORA OF NEW PROVIDENCE AND ANDROS 141

Locality. — ''Loggerhead Creek, Andros, April 22. Not common.
Said by negroes to be more common on Eleuthera and islands south."
The specimen (509) consists of two leaves and two inflorescences with
young flowers.

Inodes sp.

The genus Inodes is represented by a single inflorescence. The
calyx and corolla are longer than in /. Palmetto (I^oddiges), the calyx
more deeply lobed and the margins of the petals more distinctly papil-
late-denticulate. The two unequal triangular scales which subtend
the flowers are also somewhat larger. These facts indicate specific
distinctness, but in the absence of other data the application of a
name may be postponed.

Locality. — "In swamps along road near Nassau, February 5."
The leaf which bears the same number (284) is here referred to
Coccothrinax.

A large-leaved fan-palm from which material for weaving hats and
baskets is obtained was noticed by Mr. Lyster H. Dewey growing in
swampy places on New Providence Island. It attains a height of
about 5 m. and is possibly different from the palm reported from
Cat Island by Professor A. S. Hitchcock (Report Mo. Bot; Gar., 4,
138. 1893) as Sabal umbraculifera. This occurs in dry situations^
attains a height of about 8 m. and has an inflorescence 1.2 m. long.
The berries are 12 mm. in diameter, the seed concave at base and
the embryo lateral. As already explained elsewhere (Bull. Torrey
Club, 28, 531. 1891), the name umbraculifera was not available for
transfer from Corypha to any American palm, and the application of
the next available name Inodes Blackburniana (Glazebrook) has not
yet been fixed.

Arecace^
Cyclospatheae, new tribe

To accommodate the following genus Cyclospathe and Pseudo-
phoenix Wendland. The association of the latter with Morenia,
Gaussia, and Synechanthus as proposed by Professor Drude, seems to
have little warrant.

The Moreniaceae are monoecious palms with numerous cylindrical
partial spathes and sessile flowers arranged in rows. The Cyclo-

142 FLORA OF NEW PROVIDENCE AND ANDROS

spatheae are apparently dioecious palms with a single complete spathe
and pedicellate flowers, without orderly arrangement. The Cyclo-
spatheae may also be recognized at once by their strongly conduplicate
leaf segments, a feature in which they are especially divergent from
ChamcBdorea and the related genera, which have the bases of the seg-
ments more open than in most of the pinnate-leaved palms.

Cyclospathe gen. nov. .

A small palm obviously allied to Pseudo phoenix Wendland, but
distinct in having the trunk short and with short internodes, the
inflorescence infrafoliar, and the calyx deeply lobed ; also in the pos-
session of a curious, short, collar-like spathe completely encircling the
main axis of the spadix near the middle of its base.

The leaf segments are strongly folded together like those of Pseudo-
phoenix^ but the lower margin is not incurved to bring it against the
upper as in Pseudo phcenix. The "dark conspicuous gland-like ex-
crescences" described by Sargent (Silva, lo, 33) on the sides of the
rachis at the base of the pinnae are evidently much smaller in Cyclo-
spathe, and are mostly confined to the angle of insertion of the upper
margin of the pinnae.

Further differences between Cyclospathe and Pseudophcenix are

discussed in connection with the following description of the type

species :

Cyclospathe Northropi sp. nov.

Trunk less than 3 m. in height, about 22 cm. in diameter, slightly
bulging in the center; leaf scars distinct, about 2.5 cm. apart; leaf
bases very glaucous, also the rachis; 'rachis distally subtriangular in
section, the leaf-bases completely crossing the lateral faces and even
prominent above the narrow crest; upper and lower margins of the
pinnae inserted on the same plane at the lateral angles of the rachis ;
segments are not so strongly plicate as in Pseudophcenix, the two edges
meeting the rachis about 5 mm. apart, and not thickened and incurved
as in the heavier and larger leaf of Pseudophcenix.

The specimens studied have about 20 of the apical pinnae on each
side ; lower pinnae about 47 cm. long by 23 mm. broad ; apical pinnae
gradually reduced to 27 cm. by 10 mm. and smaller, the terminal divi-
sions not completely separated. The distal pinnae are farther apart
than the proximal.

FLORA OF NEW PROVIDENCE AND ANDROS 143

In Pseudo phoenix pinnae 43 cm. long have a width of 2)Z ^^- The
texture of the pinnae of Cyclospathe is also much thinner and more
fragile than in Pseudophosnix, and the decurved and thickened
anterior margin is broader. The upper surface shows several rather
prominent veinules not regularly spaced ; below, the veinules are very
numerous, close, and equal. The upper surface is smooth and shining,
the lower dull and uniform, the space between the veinules being
minutely roughened. In Pseudophcenix both surfaces appear more
distinctly glaucous.

The spadix at the flowering stage is about 35 cm. long, and about
12 mm. broad at the flattened base. There are nearly 20 primary
branches decreasing in size from the lowest, which is 11 cm. long and
3 mm. thick at base. The branches are twice or thrice subdivided, the
ultimate divisions being about 15 mm. long and bearing solitary flowers
at intervals of about i mm., but without regularity of arrangement
apparent in the dried specimen. -Each branch and flower is subtended
by a triangular pointed bract, those of the primary branches being 5 to
8 mm. long, strongly acuminate with a very broad base which at the
lowest fork is continued halfway round the stem. A similar growth is
probably referred. to by Sargent as a "thickened ear-like body" on the
upper side of the base of the branches of the inflorescence of Pseudo-
phosnix.

The most curious peculiarity of Cyclospathe is a further extension
of such a bract or rudimentary spathe to form a complete frill-like
band or collar about the middle of the basal stalk of the inflorescence.
This structure is about 5 mm. wide, of a light brown color; texture
firm, but thin and rather brittle in the dried state.

Flowers (perhaps immature) about 2 mm. long. Calyx tubular
forming a pedicel-like base i mm. long; at apex splitting into three
triangular slightly imbricate lobes. Corolla thick and fleshy, the petals
valvate. No stamens or staminodia were made out, and the indications
are that Cyclospathe is dioecious. Sargent characterizes Pseudophcenix
as monoecious, but apparently without reason, as he says afterward
"flowers unknown" and describes only the persistent "staminodia"
of the ripe fruit.

Locality. — Andros Island. The leaves (508) were collected on
Loggerhead Creek in April, 1890, the inflorescence (671) on "Big
Cabbage Creek, west side" in June. The local name, "hog cabbage

144 FLORA OF NEW PROVIDENCE AND ANDROS

palm," appears with both labels and increases the probability that the
specimens were properly associated. In the event of doubt on this
point the inflorescence should be treated as the type.

Cocos NUCiFERA L. Common in cultivation.

Bromeliace^

TiLLANDSiA Balbisiana R. & S. ''Wild pine." Common and
variable. Red Bays and Conch Sound, April; Lisbon Creek, June
(491, 528).

TiLLANDSiA BULBOSA Hook. "Wild onion." On mangroves in
the swash. Purser Point, June. Not common (654).

TiLLANDSiA FASCicuLATA Swartz. " Dog-drink-water." Com-
mon. Nicol's Town, Red Bays, April (439).

TiLLANDSiAF LEXUOSA Sw. Nicol's Town, March. Like Blodg-
ett's specimen from Key West; not like Wright 3271 (369).

TiLLANDSiA RECURVATA L. Not common. Fresh Creek, Kemp
Sound, Andros, June (617).

TiLLANDSiA UTRicuLATA L. More than 1.5 m. in height.
Flowers whitish. Larger than any specimens examined. Fresh Creek,
June (612).

COMMELINACE^

CoMMELiNA NUDiFLORA L. Nassau, Jan. (7).
Rhcea DISCOLOR (L'Her.) Hance. (Tradescanlia discolor L'Her.)
Nassau, Jan. ; Conch Sound, April (26).

LlLIACE^

Aletris bracteata sp. no v.

Roots numerous, fibrous; basal leaves numerous, spreading,
grayish green, lanceolate or linear-lanceolate, apex acuminate, rigid,
narrowed at the base, 6-10 cm. long, 6-10 mm. wide; scape 5-6 dm.
in height, bearing small, scattered bract-like leaves; raceme erect,
many-flowered, pedicels about i mm. in length, bracts subulate, 4-6
mm. in length, almost equaling the corolla; perianth tubular-oblong,
sometimes slightly contracted below the lobes, white, 6-S mm. long,
about 3 mm. wide, slightly roughish on the outside ; lobes six, oblong-
lanceolate, about one-fourth as long as the tube; stamens included.

Mem. Torrey Club, 12

Pl.i

ALETRIS BRACTEATA

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 145

oblong-lanceolate, apiculate at the apex, longer than the filaments;
pistil included; ovary adherent to the perianth for the lower half,
style flattened and broad at the base, slightly three-cleft above : stig-
mas three; ovules numerous; fruit not seen (463).

Common on the savannas near Red Bays, Andros, April.

Closely related to A. farinosa L., but differs in the grayish green
longer and narrower leaves, with more rigid apex, the longer bracts
and the broader flattened style.

Plate i. Aletris hracteata. Entire plant X f ; />, interior of perianth; e, stamen;
r, pistil.

SMILACACEiE

Smilax auriculata Walt. New Providence, March; Conch
Sound, April, Purser Point, June. Common. Peduncles longer than
pedicels (339, 527).

Smilax Havanensis Jacq. "Saw-brier." Nassau, Jan.; Purser
Point, Deep Creek, June.

Specimens from Nassau and Purser Point unarmed; Deep Creek
specimens have prickles on midrib of many of the leaves as well as on
margin (59, 663).

Smilax sp. ? Unarmed or with very few prickles, branchlets
angular, leaves mostly broadest at the apex, obovate or oval, 4-5 cm.
long, 3-4 cm. wide, thickish, smooth, apex retuse, mucronate, base
acute, margin entire, veins prominulous on both sides ; tendrils usually
inserted a little below the middle of the petiole: staminate flowers,
peduncles longer than the petioles, flower-buds globose, petals ellip-
tical, blunt, 3-4 mm. ; anthers oblong. Pistillate flowers not seen.
Collected on border of swash at Purser Point, Andros, June (664).

Amaryllidace^

Agave rigida Mill. "Bamboo." On savannas at Red Bays,
April. Not common. Flowers bright yellow. Determined by Mr.
J. G. Baker, at Kew% 1897.

Agave rigida Sisalana Engelm. "Sisal." Cultivated and
escaped. Nassau, Jan. (164).

FuRCRCEA Cubensis Haw. Specimens imperfect. Nassau, Jan.
(203).

146 FLORA OF NEW PROVIDENCE AND ANDROS

Hjrmenocallis arenicola sp. no v.

Bulb large ; leaves erect-spreading, fleshy, smooth, dark green,
lance-oblong, 4-5.5 dm. in length, 4-5.5 ^^- wide, rounded at the apex
and narrowing at the base to 2.5-3.5 c^- 5 scape almost equaling the
leaves; bracts large, scarious, ovate or lanceolate, 3-6 cm. long, 1-2.5
cm. wide; flowers seven to thirteen in a sessile umbel, white, fragment ;
tube filiform, white, 5-7 cm. long, shorter than the lobes ; divisions of
the perianth 8-1 1 cm., narrowly linear, recurved; crown infundibular,
3-4 cm. in length, less than half as long as the stamens, teeth prolonged
into the filaments; anthers linear, 1.5 cm. long, attached below the
middle; ovary ovate, three-celled, about 2 cm. long; style filiform,
longer than the stamens, about equaling the lobes of the perianth;
stigma small, capitate.

Common on sandy beaches on the eastern side of Ahdros. Col-
lected at Nicol's Town, April; Fresh Creek and Deep Creek, June.
Most nearly related to H. Caribcea Herb. (519).

Plate 2. Hymenocallis arenicola, inflorescence and leaf X f .

Hypoxis juncea Smith. On border of swash. Red Bays, April,
May. Not common (476). The same as Wright 3745, except that it is
smaller and more delicate.

DlOSCOREACE^

Rajania hastata L. Common on both islands. Nassau, Jan. ;
Deep Creek, June. Leaves very slender, i cm. in width, or less above
the base, otherwise like Wright, Cuba, 171 2 (203).

Orchidace^

Bletia verecunda R. Br. Common in the pines on both islands
and on the border of the swash on west side of Andros. Nassau, Jan. ;
Red Bays, March. There seems to be no distinction between this
species and B. purpurea DC, reported as endemic in the Bahamas
(91). Same as 365 Plant. Guat., J. Donnell Smith.

Epidendrum fucatum Lindl. "Wild Indian." Common in the
coppet. Conch Sound, May; Mastic Point, June; Mars Bay, July
(584, 711). Same as Wright 3329.

Epidendrum nocturnum L. Conch Sound, March. In fruit
only (414).

Mem. Torrey Club, 12

Pl. 2

-pr

HYMENOCALLIS ARENICOLA

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 147

Epidendrum odoratissimum Lindl. Growing in sandy soil, near
the shore, occasionally on trees near ground, in that case being smaller ;
flowers very fragrant, having an odor like birch. Collected at Calabash
Cay, near Stafford Creek, Andros, June (606).

Epidendrum Phceniceum Lindl. ''Wild Indian." Not uncom-
mon in the coppet. Fresh Creek, June (609). Differs from Rugel
814, in having the bracts much shorter, 5-7 mm. long, more obtuse,
and the lip more strongly crispate.

Epidendrum sp. ? Single specimen from Stafford Creek, June.
Aerial, tubers small, one-phyllous, leaf 5-6 cm. in length, linear-oblong,
crenulate : scape 3 dm. in height, few-flowered : flowers white, 2 cm.
long, divisions of perigone narrow (674).

Epidendrum? In fruit only. Aerial, tuber two-leaved, leaves
3 dm. in length, 5 mm. in width : capsule oblong, 2 cm. long, 8-10 mm.
wide. In the pines, Lisbon Creek, Andros, June (679).

Broughtonia lilacina Henfr. (Lceliopsis Domingensis Lindl.)
Common in the coppet. Cocoanut Point, April; Fresh Creek, Lisbon
Creek, June (437, 448, 546).

PoLYSTACHYA LUTEOLA Hook. Ontrces, in coppct. Conch Sound,
April (525).

PoLYSTACHYA sp. ? Closc to P. luteola, but much smaller ; leaves
12-14 cm. long, I cm. broad: flowers 5-7 mm. long, lip three-lobed,
callous at base, column very short. Conch Sound, March and May

(407).

GovENiA UTRICULATA Lindl. In fruit only. Conch Sound, March.
Determined by Mr. J. M. Greenman, at Cambridge (418).

Oncidium sylvestre Lindl. ? Not uncommon under the pines.
Conch Sound and London Creek, May. It has the habit of O. sylves-
tre, but the white pink-spotted flowers are smaller and leaves shorter ;
lip 8-10 mm. in width: leaves 8-12 cm. long (543).

Oncidium variegatum Sw. On trees. Conch Sound, May.
Leaves longer and narrower (3-4 mm. in width), than in Wright 668
and Eggers 1796 from St. Domingo (587).

Oncidium sp. On trees, rare. Fresh Creek, June. Near O.
variegatum Sw. Flowers deeply spotted, divisions of perigone nar-
rower, 2-4 mm. wide, abruptly pointed : leaves 6-8 mm. wide (647).

Oncidium sp. ? Near O. sphacelatum Lindl. Single specimen
from Mastic Point, June; terrestrial, 1.5 m. in height; scape lateral;

148 FLORA OF NEW PROVIDENCE AND ANDROS

greenish yellow flowers panicled ; leaves equitant, recurved, rosulate
from flattened tubers (602). No. 750 collected at Mars Bay, Andros,
in July is probably a smaller specimen of the same species and No, 405
from Conch Sound, an Oncidium in fruit only, may perhaps be referred
to the same species.

Vanilla articulata sp. nov. "Link-vine," "wormwood"

A tall climber with aerial roots, growing over trees and shrubs,
aphyllous ; stems jointed, joints fleshy, smooth, subangular, 2-3 dm.
in length; flowers in short axillary spikes, 6-12 flowered; bracts
broadly ovate or triangular, blunt, 5 mm. in length; flowers about 6
cm. long, fleshy, white with faint pinkish tinge, parts of perigone
jointed at the base; sepals erect, spreading, fleshy, oblanceolate,
involute at the tip, 3-4 cm. long, about i cm. wide ; petals oblanceolate
or spatulate, equaling the sepals but thinner, keeled on the back;
lip adnate to the column more than two-thirds of the way, convolute,
broadly obovate or triangular, about 3 cm. in width, channeled on the
back, three-lobed, lobes obtuse, crispate, lateral lobes papillose below,
central lobe sparingly crested above, bearded below (thick tuft of cilia
5-6 mm. long), column elongated, about 2.5 cm.; anther terminal
jointed at the base, pollinia two; stigma shortly transverse: ovary
fleshy, flattened, sometimes slightly two-edged, incurved, 3-3.5 cm.
long, 5 mm. wide; capsule elongated.

Collected on both islands, not common. New Providence, Feb. ;
in bloom, July; London Creek, May; Deep Creek, June (545).

Plate 3. Vanilla articulata. Cluster of flowers; e, lip; a, n, sepals; d, petal ; 0,
column, side view ; v, column, front view ; m, cross section of buds.

Cranichis sp. ? IsiesiTC. tenuis Reich. Leaves lanceolate, 6-8 cm.
in length; petioles equaling or exceeding the leaves; scape very
slender, few-flowered, 22 cm. in length, few small sheathing scales.
Two specimens only. Conch Sound, May (567).

Cranichis sp. ? In fruit, possibly C. muscosa Sw., but the scape is
much more densely flowered than in Wright 620 from Cuba. Conch
Sound, March (417).

Stenorhynchus orchioides Rich. Not common. Flowers
greenish. Conch Sound, May (412).

Gyrostachys Peruviana (Aubl.) Kuntze? Common and vari-
able. Collected on both sides of Andros, Conch Sound, March ; Red
Bays, April (399, 574)- Bracts more acuminate than in Wright 3296.
{Spiranthes tortilis Rich. ?)

Mem. Torrey Club, 12

-/\f^

VANILLA ARTIENLATA

HELIOTYPE CO., 30ST0N.

FLORA OF NEW PROVIDENCE AND ANDROS 149

LiMODORUM TUBEROSUM L. Coromon on savannas on west side

of Andros. Variable, some specimens approach L. graminifolium (Ell.)

Small. Red Bays, April, May (430-500). Same as Wright 3317 from

Cuba.

Casuarinace^

Casuarina equisetifolia Forst. ''Spanish cedar." Nassau,

Feb. Common in cultivation (297, 454). 454 was collected on the

west shore of Andros, miles from any settlement. It is also reported

from the Florida Keys.

Myricace^

Myrica cerifera L. "Wax-berry," "mickle-berry." Common
on Andros. Nicol's Town, March; Lisbon Creek, July (357).

MORACEiE

Ficus DiMiDiATA Griscb. "Fig tree." Nassau, Jan.; Nicol's
^Town, March (119, 377, 378). 119 is the same as Wright 542. 377
and 378 are probably F. dimidiata with young leaves.

Ficus pedunculata Willd. "Fig tree." Nassau, Jan. (46).
Same as Wright 1684.

Ficus pertusa L. Mastic Point, Andros, May (586). Same as
Wright 545.

Ficus Indica L. "Banyan." Cultivated, Nassau, Jan. (295).
Artocarpus incisa L. f. "Breadfruit." Cultivated (261).

Ulmace^

Trema Lima (Lam.) A. S. Hitch. ? {Sponia Lamarckiana Desc.)
"Wild birch," "wild fig." Common in the coppet at Red Bays
(Lewis Coppet), April; Deep Creek, June.

A low tree, not tortuous branching. In general appearance and
mode of branching seems between T. mollis Desc. and T. Lima. The
leaves are larger than any specimen of T. Lima examined, 4-5 cm.
long, .5-2.5 cm. broad; upper surface very scabrous, apex acute. It
is the same as Cooper 21 from New Providence and Eggers 2326, un-
named (485, 683).

URTICACEiE

Fleurya tESTuans Gaud. Nassau, Jan. (30).

Adicea microphylla (Sw.) Kuntze. Nassau, Jan. (29).

150 FLORA OF NEW PROVIDENCE AND ANDROS

LORANTHACEiE

Dendropemon emarginatus (Sw.) Steud. Nicol's Town, March.
On fig, gum elemi, etc. (373). Agrees with Wright 1303, except that
racemes are shorter and the pedicels longer.

Dendropemon sp. "Mistletoe." Mars Bay, Andros; on Pelto-
phorum, July (713). Plants smooth, branchlets and peduncles much
compressed.

Phoradendron racemosum Krug & Urban. '' Snake-root," " big
man." Deep Creek, June; on Nectandra (704). Same as Eggers
1 741 and Wright 12^2 p. p.

Phoradendron Northropiae Urban, sp. nov.

Ramis teretibus v. junioribus plus minus compressis, superne di-v.
-trichotomis : vaginis cataphyllaribus ad omnia internodia supra basin
obviis solitariis, raro binis : foliis obovatis v. breviter obovatis antice
rotundatis v. sub-truncatis, plerumque late emarginatis, basi sensim
V. satis abrupte in petiolum 2-6 mm. longum angustatis, 3-7 cm. longis,
2.5-4 cm. latis, vix dimidio usque duplo longioribus quam latioribus,
partissime et obsolete pinnatinervibus, crasse coriaceis; spicis ad no-
dos pluribus lateralibus, 1.5-2.5 cm. longis; 4-6 articulatis: articulis
androgynis, 10-14 floris v. 1-2 supremis 6-2-floris; floribus in seriebus
4 depositis, imparibus 2 sub apice cujusque articuli adjectis, hisve
masculis, caeteris femineis, baccis non visis.

Rami inf erne 3-5 mm. crassi, glaberrimi, internodiis 4-10 cm. longis.
Folia in sicco olivacea v. brunnescentia, nervis lateralibus utrinque
plerumque 2, altero supra basin, altero ad medium e nervo medio
abeunte, supra vix, subtus paullo melius conspicuo. Spicae interdum
revera ex axillis euphyllorum solitariae, sed utrinque accessoriis autae
ideoque pro axilla specie ternae, sed plerumque ad nodos vetustos
inordinate plures, ex axillis squamarum minutarum orientes, 3-5 mm.
longe pedunculatas.

Hab. in Ins. Bahamas, Andros Island and Conch Sound and Lisbon
Creek, in June flor. Northrop No. 551 (in Mimusops depressa Pierre).

Plate 4. Phoradendron Northropice. Portion of plant X f ; a, inflorescence.
ARISTOLOCHIACEiE

Aristolochia passiflor^folia Rich. Conch Sound, May.
In the pines (568). Same as Wright 3665.

Aristolochia pentandra L. Nicol's Town, March (385).
Same as specimens of Garber and of Curtiss from South Florida.

EM. ToRREY Club, 12

Pl. 4

PHORODENDRON NORTHROPIAE

HEUOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 151

POLYGONACE^

CoccoLOBis DiVERSiFOLiA Jacq. Nassau, Jan. (143). Deter-
mined at Kew by Dr. N. L. Britton. Same as Brace 142 "var. foliis
minoribus" Lindau.

CoccoLOBis RETUSA Griseb. ''Pigeon-plum." Deep Creek,
June (705) ; Purser Point, June (662) ; low in fruit. Same as Wright
2252.

COCCOLOBIS TENUiFOLiA L. Nicol's Town, April (443). Deter-
mined at Kew by Dr. N. L. Britton. Same as Wright 3368. Same as
Brace 151, 193, and 205.

COCCOLOBIS uviFER^ (L.) Jacq. ''Sea-grape." Nassau, Jan.
Common everywhere along north shore of New Providence and east
shore of Andros (80).

COCCOLOBIS Wrightii Lindau. "Pigeon-plum." Deep Creek,
July (721). Determined at Kew by Dr. N. L. Britton. Same as
Wright 1395.

COCCOLOBIS pBTUSiFOLiA Jacq. "Pigcon-plum." Deep Creek,
June (706). Determined at Kew by Dr. Britton to be the same as
Eggers 4486, which according to Lindau is C. microstachya ovalifolia
Meisn.

Polygonum Portoricense Bertero. (P. densiflorum Meisn.)
Fresh Creek, June. Not common (621).

Antigonum leptopus H. & A. Common in cultivation. Nassau
and Mastic Point, June (604).

Chenopodiace^

Atriplex cristata H. B. K. Deep Creek, June (709).

Salicornia ambigua Michx. Common along the east coast.
Nassau, Jan. (171).

Salicornia Bigelovii Torr. (5. mucronala Bigel.) In the
swash on "the west side of Andros. Wide Opening, June (669).
3 dm. in height, branches very strict.

DoNDiA FRUTICOSA (Forsk.) (SucBda.) Red Bays, April (455).
7-8 dm. in height, rigid, much branched.

DoNDiA linearis (Ell.) Millsp. Common. Nassau, Jan. (194).

152 FLORA OF NEW PROVIDENCE AND ANDROS

Amarantace^

Alternanthera muscoides Sw. Lake Waterloo, Nassau, Jan.

(15°)-

Alternanthera paronchyioides St. Hil. Nassau, Jan. (197).

Same as Eggers 2571 (not named specifically), except that the plant is

more compact and the petioles shorter, 3-4 mm.

LiTHOPHiLA VERMicuLARis (L.) Uline. " Sampire." Ft. Mon-
tagu, Nassau, Jan. Common on sandy shores (147).

Ire SINE PANicuLATA (L.) Kuntze. (/. celosioides L.) " Newburn
weed." Nicol's Town, March; Stafford Creek, June (362).

Batide^

Batis maritima L. Mastic Point, May; Cormorant Cay, Andros,
June (595).

Phytolaccace^

Rivina humilis glabra L. Common on N.P. and Andros.
Nassau, Jan. (20).

Phytolacca octandra L. " Poke-bush." Nicol's Town, March

(354).

Nyctaginace^

Mirabilis Jalapa L. " Four-o'clock." Naturalized in Nassau,
Jan.

BoERHAAViA ERECTA L. Near caves. New Providence, Feb. (280).

BoERHAAViA PANICULATA Rich. Mastic Point, Andros, May (732).

BoERHAAViA SCANDENS L. Near caves. New Providence, Feb.
(281).

PisoNiA ACULEATA L. '' Cockspur." Common in the coppet of
both islands. Nassau, Jan. (73, 196).

PisoNiA OBTUSATA Sw. In the coppet, Nicol's Town, April (517).
Same as a specimen of Blodgett's from Key West, named by Torrey.

PisoNiA ROTUNDATA Griscb. In the coppet, not common. Fresh
Creek, June (636). Same as Wright 3369, also specimens of Blodgett's
from Pine Key, Fla.

BouGAiNViLLEA SPECTABILIS Poir. Common in cultivation at
Nassau (142).

FLORA OF NEW PROVIDENCE AND ANDROS 153

AlZOACE^

Sesuvium portulacastrum L. Common on sandy beaches on
both islands, variable. Nassau, Jan. (149).

PORTULACACE^

PoRTULACA OLERACEA L. Common prostrate form, collected at
Nassau, Jan.

PoRTULACA OLERACEA L. var. ? Plants 2-2.5 ^^' high, ascending,
nearly erect; leaves 1-2.25 cm. in length, obtuse or sometimes retuse,
axils shortly and sparsely pilose : flowers larger than in the common
form, clustered : petals 4-5 mm. : sepals strongly carinate-winged :
style 3-4-parted: seeds small, less than .5 mm., rugose. Found
growing abundantly on Cormorant Cay in the Northern Bight, west
side of Andros, June (658).

PoRTULACA HALiMOiDES L. Conch Sound, May. Common on
rocks (580).

ANONACE.E

Anona palustris L. "Custard apple." Not uncommon in
swampy parts of the coppet. Conch Sound, March (408). Same as
Rugel 710.

Anona squamosa L. "Sugar apple." Nicol's Town, cultivated

(513)-

Ranunculace^

Clematis dioica L. Collected by Mr. Alexander Keith at Conch
Sound, 1894. Not common (742).

Laurace^

Nectandra sanguinea Rottb. "Sweet torchwood." Common
in the coppet. Conch Sound, Red Bays, April; Fresh Creek, June
(487, 613). Agrees with Wright 484 except that the leaves are narrower
(2.5-3 cm.).

Cassytha filiformis L. Common on shrubs and low trees.
Nassau, Jan. (78, 104, 512). A more slender form with yellow stems
was found on low grasses at Loggerhead Creek on the west side of
Andros, also found on Rose Island, N.P. (266).

154 FLORA OF NEW PROVIDENCE AND ANDROS

Persea Persea (L.) Cockerell. "Avocado pear." Common in
cultivation. Nicol's Town, March (371).

Papaverace^
Argemone Mexicana L. Common about Nassau, Jan. (51).

Crucifer^

Brassica arvensis (L.) B. S. P. {B. Sinapistrum Boiss). Nassau,
Jan. (225). Leaves almost entire.

Lepidium Virginicum L. Nassau, Jan. (140).

Cakile ^qualis L'Her. Common on sandy beaches. Salt
Cay, Feb. {%^, 278). Same as Wright 1863.

Capparidace^

Pedicellaria pentaphylla (L.) Schrank. ''Wild mustard."
Fresh Creek, June (630).

Crassulace^

Bryophyllum pinnatum (Lam.) S. Kurz. {B. calycinum
Salisb.) "Live-for-ever." Nassau, Jan.; also on Andros (199).

Rosacea

Chrysobalanus IcAco L. "Pigeon plum." Common on sandy
shores. Nassau, Jan. ; Fresh Creek, June. Many specimens at latter
locality had white drupes (115).

MiMOSACEiE

Acacia choriophylla Benth. "Cinnecord." Common in the
pines on both islands. New Providence, Feb. ; Nicol's Town, March
(312, 364).

Acacia Farnesiana (L.) Willd. Nassau, Jan. (44).

AcuAN DEPRESSA (Kunth.) Kuntzc. (Desmanihus.) Red Bays,
April (495).

AcuANViRGATA (Willd.) Kuntzc. {Desmanthus.) Mastic Point,
May (731).

Mimosa pudica L. "Sensitive plant." Collected by Mr. Keith,
1891. Uncommon on Andros, but said to be abundant on Eleuthera.
Mastic Point (735).

Mem. Torrey Club, 12

Pl.5

PITHECOLOBIUM BAHAMENSE

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 155

Leuc^na glauca (L.) Benth. " Jumby-bean." Nassau, Jan.,
Feb.; Bearing Point, Andros, June (43, 282, 657). Nos. 282 and 657
were shrubs, 1.5-2 m. in height.

Lysiloma PAUCiroLiOLA (DC.) A. S. Hitch. (Sabicu Benth.)
^'Horseflesh," "sabicu." Common in the coppet; one of the most
valuable timber trees. Nicol's Town, April; Deep Creek, June (434).

Lysiloma Bahamensis Benth. "Wild tamarind." New Provi-
dence, Feb. (309). Same as Wright 3542.

Calliandra FORMOSA Benth. Hog Island, Nassau, Feb. (255).
In fruit only. Determined by Dr. B. L. Robinson at Gray herbarium.

PiTHECOLOBiUM Hystrix Benth. (P. calliandrcBfolium Wright.)
In the pines, not common. Conch Sound, May; Deep Creek, June
(575). It is the same as Wright 2401.

PiTHECOLOBiUM Unguis-cati (L.) Benth. "Ram's horn."
Common near the shore on both islands. Nassau, Jan. ; Cocoanut
Point ; Andros, April (66, 235, 449). 66, collected at Nassau, has the
leaves two-jugal and agrees with Eggers (3830).

Pithecolobium Bahamense sp. nov. "Ram's horn"

A shrub, 1.5-2 m. in height, with slender, drooping branches,
armed with delicate, stipular spines, 3-7 mm. in length ; leaves bipin-
nate, pinnae one-jugal, leaflets one-jugal ; petiole 1-4 mm. in length,
petiolules 1-3 mm., channeled and with a single stipitate gland at the
base, leaflets oblong to oblanceolate or obovate, 1.25-2 cm. long, a little
over half as broad, obtuse or mucronate at the apex, oblique and slightly
unequal at the base, chartaceous, glabrous, shining above, entire, sub-
sessile with a stipitate gland at base ; inflorescence capitate, many-
flowered; peduncles 2-3 cm. in length, much exceeding the leaves;
calyx tubular, less than one-half the length of the corolla with five ovate
acute lobes; corolla crimson, tubular, 3-5 mm. in length, five-lobed,
lobes acute, about one-half as long as the tube; stamens numerous,
crimson, exserted, more than twice the length of the corolla; anthers
small, rounded ; ovary stipitate : style much exserted, about 2 cm. in
length, four times as long as the corolla ; stigma small ; immature fruit
compressed, slightly curved, puberulent; mature fruit 9 cm. in length,
I cm. in width, dark brown, curved ; seeds ariled.

Collected in fruit near Nassau in January, and in flower in the
coppet at Mastic Point, Andros, in June (605). The plant belongs to
the section Unguis-cati and is related to P. circinale Benth.

Plate 5. Pithecolobium Bahamense. Portions of plant in flower and in immature
fruit. X f .

156 FLORA OF NEW PROVIDENCE AND ANDROS

C^SALPINACE^

Cassia Bahamensis Mill. ''Stinking pea." New Providence
Jan. (103).

Cassia biflora L. Nassau, Jan. Leaflets strongly emarginate

(55)-

Cassia ligustrina L. Common. Nassau, Jan. ; Conch Sound,

March (123, 422). Same as Wright 1190.

Cassia mimosoides L. Nassau, Jan. Pods 2-2.5 cm. long, very
hairy (134).

Cassia occidentalis L. Nassau, Jan. (105).

Cassia polyadena DC. Nicol's Town, March. In the pines,
strict and unbranched, 7-9 cm. in height, not as pubescent as Wright
2376.

Cassia villosa Mill. Nassau, Jan. Determined by Miss Anna
Murray Vail at Columbia herbarium (14).

Cassia Caribaja sp. nov.

Shrubby, 6-9 dm. in height, stem gray, smooth; leaves 2-3.5 cm.
long, with from two to four pairs of leaflets, mostly three ; petiole 2-5
mm. long, channeled ; a stipitate gland between the leaflets or slightly
below (often wanting between the lowest pair) ; leaflets sessile, eUip-
tical, 1.5-2 cm. long, about one-third as wide, unequal and oblique at
the base, apex mucronate, margin entire, thickish, glabrous, shining
above, resinous-dotted below ; veins numerous, parallel and prominent ;
stipules about 3 mm. in length, subulate, subspinescent, ribbed and
often with appressed white hairs on the margin, persistent; flowers
large, solitary, axillary; peduncle 2-3 cm. long; calyx deciduous:
sepals lanceolate, acuminate, about i cm. long, the three outer ones
keeled, keel pilose, petals yellow, oblanceolate or obovate, nearly
twice as long as the sepals and about i cm. in width ; stamens 10, the
three upper difform, anthers linear, puberulous along the furrow,
bursting at the top by two short clefts ; ovary about 5 mm. in length,
shorter than the flattened style and covered with appressed white
hairs; legume purplish brown, linear-oblong, 3-5.5 cm. long, 5-7 mm.
wide, bivalved, compressed, mucronate, with a few scattered hairs:
seeds about eight, oblong, compressed.

Collected in the coppet at Fresh Creek, Andros, June 10, most
closely related to C. lineata Sw. (6^S).

Plate 6. Cassia Caribcsa. Portions of plant, natural size; i, leaflet.

Mem. Torrey Club, i3

Pl. 6

CASSIA CARIBAEA

HELIOTYPE CO., BOSTON,

FLORA OF NEW PROVIDENCE AND ANDROS 157

Tamarindus Indica L. " Tamarind." Fresh Creek, June ; Deep
Creek, July (642, 717).

H^matoxylon Campechianum L. "Logwood." Nassau, Jan. ;
Lisbon Creek, Andros. Naturalized (195).

C^SALPiNiA OVALIFOLIA Urban. "Nicker-bean." Nassau, Jan.
(116). Described in SymbolaeAntillanse, 2273. O. 1900.

C^SALPiNiA CRISTA L. (C. Bofiducella L.) "Nicker-tree."
Nassau, Jan. (106). Both species quite common on New Providence
and Andros on sandy beaches.

C^SALPiNiA RuGELiANA Urban. Common in the pines. Conch
Sound, March (426). Same as Wright 2364.

Peltophorum ADNATUM Griseb. "Horse-bush." Upper surface
of leaflets hirsute. Otherwise same as Wright 2359. Deep Creek,
July (712).

PoiNCiANA REGIA Boj. "Poinciana," "flamboyant." Cultivated
at Nassau, Mastic Point and Deep Creek, Andros (591).

C^SALPiNiA PULCHERRiMA Sw. " Pride of Barbadoes." Cultivated
at Nassau and Mastic Point (10).

Papilionace^

SoPHORA TOMENTOSA L. Conch Sound, March; Deep Creek,

July (411, 727). Same as Eggers 2573, but is not nearly as tomentose.

Crotalaria pumila Ort. Nassau, Jan. (67). Same as Curtiss

533-

Crotalaria retusa L. Nassau, Jan. 221. Same as Wright 117.

Crotalaria VERRUCOSA L. Nassau, Jan. (54).

Indigofera Anil L. Nassau, Jan. (175).

Cracca Schottii Vail. In old field, Lisbon Creek, Andros, June
(678). Agrees with specimens from Cartagena, Schott, No. 16. Deter-
mined by Miss Anna Murray Vail.

Stylosanthes hamata (L.) Taub. (5. procumhens Sw.) Nassau,
Jan. (37).

Meibomia INC ana (Sw.) Kuntze. (Desmodium incanum (Sw.)
DC.) Nassau, Jan. Common (224).

Abrus precatorius L. "Crab's eyes," "Black-eyed-Susan,"
"Wild licorice." Nassau, Jan. (239).

Bradburya Virginiana (L.) Kuntze. {Centrosema.) Very com-

158 FLORA OF NEW PROVIDENCE AND ANDROS

mon, fruit and leaves variable. Nassau, Jan. ; Fresh Creek, June
(82, 227, 629).

Bradburya Virginiana angustifolia (DC.) Griseb. Fresh
Creek, June (756).

Galactia rudolphioides (Griseb.) Wright. Common. Nas-
sau, Jan. ; Nicol's Town, March, April (214, 219, 395). Agrees with
Wright 1 181.

Galactia Cubensis H. B. K. (G. spiciformis T. & G.) Nas-
sau, Jan.; Fresh Creek, June (60, 226, 648). Agrees with
Wright 2332.

Canavalia OBTUSiFOLiA (Lam.) DC. ''Horse-bean." Common
on sandy beaches. Nassau, Jan.; Conch Sound, April (112, 452).
Legume 2.5 cm. wide. Same as Curtiss 682, and Eggers 2724,
unnamed specifically.

DoLiCHOLUS MINIMUS (L.) Medic. Nassau, Jan. (215). Same
as Curtiss 656.

Phaseolus semierectus L. Common. Southwest Beach, N.P.,
Jan.; Mangrove Cay, Andros, June (329). Same as Wright 137.

ViGNA REPENS (L.) Kuntzc. Nicol's Town, March (386). Same
as J. Donnell Smith 181, PL Guat.

Cajanus Cajan (L.) Millsp. "Pigeon-pea." Cultivated and
escaped on both islands. Nassau, Jan. (98).

Dalbergia Brownei (Pers.) Kuntze. "Ti-ti." Kemp Sound,
Andros, June (680).

IcHTHYOMETHiA PisciPULA (L.) A. S. Hitchc. (Pisctdia
Erythrina L.) Conch Sound, May (588). Same as Curtiss 685.

Arachis hypog^a L. "Groundnut." Cultivated. Mastic
Point, May (600).

Erythrina Corallodendron L. "Lightning tree." Culti-
vated. Nassau (308).

OXALIDACE^

OxALis CORNICULATA REPENS Zucc. Nassau, Feb. (330).

OxALis sp. Acaulescent, bulbs scaly; leaflets broadly obcor-
date, about 5 cm. wide and 2 cm. long; petioles 14-16 cm., a little
longer than the scape: flowers pale purple, about i cm. wide.
Near O. latifolia Kunth. Nassau, Feb. Probably escaped (331).

Mem. Torrey Club, 12

• 7

LINUM BAHAMENSE

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 159

Linages
Linum Bahamense sp. nov.

Suffruticose, 3-6 dm. in height, corymbosely branched, branches
erect- ascending ; stems sulcate; leaves numerous, alternate, sessile,
appressed-ascending, whitish, linear-lanceolate, 8-1 1 mm. long, a
little more than i mm. wide, one-nerved, the midrib prominent on
the under side and projecting at the apex so as to form a mucro,
margin entire or the younger leaves glandular-ciliate, glabrous or
with a few scattered hairs at the base or along the midrib on the
upper side; two dark stipular glands at the base of the leaves;
flowers corymbose, numerous; pedicels short, 2 mm. in length,
bracts ciliate-glandular ; sepals 5, lanceolate, acuminate, 2-3 mm.
long, ciliate-glandular, keeled, persistent; petals bright yellow, more
than twice the length of the sepals, obovate; stamens 5, united at
the base included; anthers oblong; styles 5, distinct, filiform; ovary
globose, imperfectly lo-celled, ovules 10, capsule globose, 2-3.5 ^"^•
in diameter, about as long as the calyx, splitting into 10 valves; seeds
oblong, compressed, reddish-brown (204, 496).

Common in the pines on both islands. Collected at Lake
Waterloo near Nassau, Jan. 25; near Southwest Beach, N.P., Feb.
26; at Red Bays on the west side of Andros, April 17; also col-
lected in N.P. by Eggers, No. 4181.

This is related to L. sulcatum Riddell, but the styles are dis-
tinct, the leaves are one-nerved and the septa of the capsule are
not glandular.

Plate 7. Linum Bahamense. Entire plant, X f ; /, flowers and buds ; r, sepal ;
a, petal; s, stamen; u, anther; p, fruit.

Erythroxylon brevipes DC. "Rat- wood." Deep Creek,
June (692). Same as Wright 2134 and Eggers 2435 unnamed
specifically.

Erythroxylon obovatum Macf. Mastic Point, May (597).
Same as Eggers 4345 from Bahama; leaves thicker and veining
more conspicuous than in Wright 2140 from Cuba.

Erythroxylon reticulatum sp. nov.

Tall shrub with slender branches; reddish brown verrucose
bark, branchlets strongly compressed; leaves oblanceolate or
oblong, 2.5-3 cm. long, half as wide, broadly obtuse and mucron-
ulate at the apex, narrowed below to a short petiole, 2-3 mm.

160 FLORA OF NEW PROVIDENCE AND ANDROS

long, glabrous, thickish, entire, dull green above, pale below, are-
olate, the reddish connecting veins circumscribing a central area
4-6 mm. wide; midrib reddish, prominent, veins delicate, promin-
ulous above; stipules persistent, triangular, acuminate, shorter
than the petioles, reddish brown; flowers axillary, appearing with
the leaves, solitary or sometimes in pairs; pedicels 5-8 mm. long,
slender below, gradually thickening and wing-angled above; calyx
spreading, sepals 5, lanceolate, about one- third the length of the
petals; petals white, 3-3.5 mm. in length, deciduous, elliptical-ob-
long, slightly keeled on the back, internal scale 2-lobed and con-
torted at the apex; stamens 10, 4-5 mm. long, exserted, filaments
united over one-third of the way, tube extending a little beyond
the sepals; ovary oblong, 3-celled; styles 3, distinct; stigmas flat-
tened, reddish; young drupe oblong, pointed, purplish black, 5
mm. in length.

Collected at Deep Creek, Andros, June 27; growing in sand
(682). Most nearly related to E. areolatum L.

Plate 8. Erythroxylon reticulatum. Portion of plant, xf ; a, flowering branch;
d, flower without corolla ; w, petal.

Malpighiace^

Byrsonima lucida Rich. Common on Andros on the edge of
the coppet. Our specimens resemble those from Florida more
closely than they do the West Indian specimens; the latter all have
broader, and obovate, instead of oblanceolate or spatulate leaves.
Nicol's Town, March (367). Same as Curtiss 501. Largest speci-
men seen one foot in diameter and about twenty feet in height.

Malpighia setosa Spr. "Touch-me-not." Fresh Creek,
June (737). Determined by Dr. Britton at Kew to be "the same
as a specimen marked M. setosa by Jussieu; collected also by Brace
No. 114."

A number of specimens were collected at various times which
seem to vary greatly; possibly more than one species is repre-
sented. No. 737 was coflected in the pines at Fresh Creek. It
was a tall shrub with whitish bark, oblong leaves, 2-3.5 cm.
long, 12-20 mm. wide, entire or nearly so, hairs few on mar-
gin and under surface; inflorescence, two-flowered umbels; pedi-
cels 8-13 mm.; flowers rosy, 12-15 "^^- wide; drupe globose,
furrowed.

Mem. Torrey Club, 12

ERYTHROXYLON RETIENLATUM

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 161

No. 538 was a single specimen from the pines at Stafford
Creek, collected in May, differing from the above in having the
leaves thicker but only half as wide, 2-3 cm. long and 8-10 mm.
broad, pointed at base and apex, margin dentate or strongly den-
ticulate. No. 538a is a specimen in fruit, collected at Conch
Sound, leaves broader than in above, denticulate and with hairs
on both sides of the leaf; 5386, also from Conch Sound, sent by
Mr. Keith, has oval or ovate leaves 2-2.5 c^- wide, with strongly
dentate margins, hairs beneath and on the margin, flowers smaller
than in 737. All the above specimens differ very much from Wr.
No. 99, marked M. setosa, which is in leaf only; that has leaves less
than half the size of any of ours, hairy above and strongly hirsute be-
neath.

Stigmatophyllon Sagr^num Juss. Not uncommon in the
pines. Conch Sound, March; Red Bays, April (401-468). Same
as Rugel 157.

Triopteris rigida Sw. Common in the coppet. Nassau,
Feb.; Conch Sound, May; Calabash Cay, Andros, June (217).
Same as Wright 96. 217a from Deep Creek has the leaves oblance-
olate or obovate, 12-17 ^^- wide; all the other specimens col-
lected have oblong or elliptical leaves 7-10 mm. wide.

RUTAC^E

Fagara flava (Vahl) Krug & Urban. (Xanthoxylon crihrosum
Spreng.) "Satin-wood." Deep Creek, June (695)..

Fagara coriacea (A. Rich) Krug & Urban. {Xanthoxylon
emarginatum Wright.) "Hercules' club." Common in the coppet.
Nicol'sTown, March; Mastic Point, April; Deep Creek, June (372).
Same as Eggers 4034, 4503, Brace 510. Carpels globose, not
oblique.

Fagara Fagara (L.) Small. (Fagara lentiscifolia Willd.)
Nassau, Nicol's Town (747).

SlMARUBE^

SuRiANA MARiTiMA L. "Bay-ccdar." Common along the
shore on both islands. New Providence, Jan. (86). Same as
Eggers 2728.

PiCRODENDRON BACCATUM Bahamense Krug & Urban. Conch

162 FLORA OF NEW PROVIDENCE AND ANDROS

Sound, April (453). Determined at Kew by Dr. Britton. Same
as Eggers 4402, Brace 476.

BURSERACE^

BuRSERA SiMARUBA (L.) Sarg. "Gum elemi," "West Indian
birch." Common on both islands. Nassau, Jan.; Conch Sound,
May (64, 560).

SwiETENiA Mahogani L. "Madeira." Nassau, Jan.; Man-
grove Cay, June (137, 676). Same as Eggers 1836, Wright

1153-

POLYGALACE^

PoLYGALA BoYKiNii Nutt. Common on savaimas on west
side of Andros. Red Bays, April (473).

PoLYGALA BRizoiDES St. Hil. Common in savannas on west
side of Andros. Red Bays, April (465). Determined at Kew
by Mr. A. W. Bennett — "sed racemi quandoque axillares."

PoLYGALA SPATHULATA Griscb. Conch Sound, March (402).
Determined by Mr. A. W. Bennett at Kew.

EUPHORBIACE^

Buxus Bahamensis Baker. On savanna. Near the shore at
Red Bays, April (460). Determined by Dr. B. L. Robinson at
Gray herbarium.

Phyllanthus Epiphyllanthus L. (P. falcatus Sw.) "Hard-
head." Conmion on both islands. Nassau, Feb.; Nicol's Town,
March (146-325). Same as Eggers 7234. Branchlets narrower
than in Wright 1951. No. 146a and 146& have branchlets very
narrow, 8-10 cm. long and 6-8 mm. wide, apex more acute,
pedicels longer and more slender, 3 mm. in length.

Phyllanthus distichus L. "Gooseberry tree.'' Fresh
Creek, June, escaped (653).

Phyllanthus Niruri L. Nicol's Town, March (338).

Phyllanthus Bahamensis Urb. Lewis Coppet near Red
Bays, Andros, May (488). Same as Eggers 4241, 4464.

Savia erythroxyloides Griseb. "Maiden-bush." Not un-
frequent in the coppet. Fresh Creek, Deep Creek, June. In fruit
only (610). Determined by Mr. M. L. Fernald.

Croton linearis Jacq. (C. Cascarilla linearis Jacq.) " Granny-

FLORA OF NEW PROVIDENCE AND ANDROS 163

bush." Common along the eastern shores of both islands. Nassau,
Jan.; Fresh Creek, June (113, 615).

Manihot Manihot (L.) Cockerell. {Jatropha Manihot L.)
"Sweet cassava." Cultivated and escaped. Nicol's Town, March

(363)-

AcALYPHA ALOPECUROiDES Jacq. Nassau, Jan. (32). Same as
Wright 571.

Lasiocroton macrophyllus Griseb. "Wild oak," "Light-
wood," "Bitters." Deep Creek, June (689). Agrees with speci-
men of March from Jamaica except that the upper surface of the
leaves is smooth instead of velvety, and the apex is obtuse instead of
acute.

ExccECARiA LUCiDA Sw. "Crab-wood." Nicol's Town, March;
Fresh Creek, June (375). '

ExccECARiA Sagr^i J. Mlill. Stafford Creek, Andros, May
(589). Same as Wright 2006.

HipPOMANE Mancinella L. " Manchinecl." Not uncommon
on Andros. Conch Sound, May; Fresh Creek, June (556, 622).
In 662 the sap did not seem at all milky. All the herbarium specimens
examined showed the leaf apex acute or acuminate, while all of ours
have the apex very blunt.

BoNANiA EMARGiNATA DC. Fresh Creek, June (628).

Pedilanthus sp. ? possibly P. angustifolius Poit.

Shrubby, 7-9 cm. high, half scandent with green, rather fleshy
stems, very little branched. All the plants seen were destitute of
leaves though alternate- leaf scars are discernible, 4-5 cm. apart,
scars of [the floral leaves opposite; flowers in terminal cymes, pe-
duncles 4-5 cm. in length; involucre bright red, pubescent, about
8 mm. long, irregular, with a sharp spur at the side, 5 mm. long;
spurred part of involucre with four glands at base : staminate flowers
indefinite : pistillate ones exserted : style 10 mm. in length.

Collected in two localities in the coppet at Deep Creek, June

(693)-

Resembles P. angustifolius Poit. as shown by 769 PL Sintensis
in Gray herbarium, but specimen not in good flower.

Euphorbia Blodgettii Engelm. Nassau, Jan. ; Nicol's Town,
March (41, 379).

Euphorbia buxifolia Lam. Common on sandy beaches.

164 FLORA OF NEW PROVIDENCE AND ANDROS

Southwest Beach, N.P., Jan. ; Red Bays, April (87, 457). No. 672,
a form with leaves not appressed, was collected at Big Cabbage
Creek on the west side of Andros in June. Same as Wright 2016.

Euphorbia cassythoides Boiss. Rare, Deep Creek, June
(702). Determined at Kew, 1897.

Euphorbia heterophylla L.- Common in the pines. New
Providence, Jan. (96).

Euphorbia heterophylla graminifolia Engelm. Same lo-
cality (92).

Euphorbia nutans Lag. Common. Nassau, Jan.; NicoPs
Town, March (229, 380). Our specimens agree with Professor
Hitchcock's in having the upper internodes slender, especially in
229, and no dark spots on the leaves. (See Hitchcock's Report,
Plants of the Bahamas, Jamaica and Grand Cayman.)

Euphorbia serpens Kunth. Nassau, Jan. (277). Same as
specimen of Rugel, No. 148 marked E. serpens var. ? Much branched,
delicate, with slender wiry stems, enlarged nodes, leaves 3-4 mm. in
length.

Euphorbia PULCHERRiMA (Graham) Boiss. "Poinsettia." Nas-
sau, common in cultivation.

Euphorbia antiquorum L. Nassau, cultivated.

HuRA crepitans L. "Sand-box tree." Nassau, cultivated.

ANACARDIACEiE

Metopium Metopium (L.) Small. "Poison- wood." Com-
mon in the coppet on both islands. Nassau, Jan.; Conch Sound,
May (70, 552). Same as Wright 2287, Curtiss 448.

MORINGACE^

MoRiNGA MoRiNGA (L.) Small. "Horse-radish tree." Nassau,
cultivated (310).

Celastrace^e

El^odendrum xylocarpum DC. Nassau, Jan. (71). In
fruit only. Determined by Mr. M. L. Fernald, Cambridge.

Maytenus buxifolius (Rich.) Griseb. Fresh Creek, June.
Differs from Wright 2215 and all other specimens examined in having

Mem. Torrey Club, 12

Pl. 9

CROSSOPETALUM CORIACEUM

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 165

the leaves narrower and more spatulate, 2-2.5 cm. long, 6-9 mm.
wide (626).

Crossopetalum fallens (Smith). (Myginda pallens Smith.)
Cocoanut Point, April; Deep Creek, June (447, 718). Like Eggers
4140 and 4438 from New Providence, ''put with M. pallens at Kew."
Differs from the Florida specimens in having much narrower leaves,
oblanceolate or spatulate, 2-2.5 cm. long, 4-9 mm. wide.

Crossopetalum coriaceum sp. nov.

Low shrub; branches ascending, bark grayish, branchlets
tetragonal, ends somewhat wing-angled; leaves numerous, opposite,
subsessile, coriaceous, broadly oblanceolate, 1.5-2 cm. long, about
I cm. wide, apex obtuse or sometimes slightly retuse, base cuneate,
margin entire, somewhat revolute, veins inconspicuous; flowers
minute, reddish, in axillary subsessile cymes, 2-5-flowered, pedicels
1-2 mm. in length, jointed, pedicels and calyx puberulous; calyx
campanulate, persistent, lobes 4, rounded, reddish, obtuse; petals 4,
orbicular, longer than the calyx, spreading or at length reflexed;
stamens 4, inserted on the edge of the disk; filaments subulate;
anthers small, globose; ovary immersed in the disk, globose: style
short : stigmas 4 : fruit a red drupe, slightly obovate, about 3 mm.
in length, one-celled, one-seeded (480).

Plate 9. Crossopetalum coriaceum. Portion of plant X f ; j-, stamens ; w, ovary ;
a, fruit.

Collected on the savannas at Red Bays. Andros, April. Allied
to C. pallens (Smith), but differs in having thicker entire leaves and
smaller subsessile cymes.

ScH^FFERiA FRUTESCENS Jacq. Deep Creek, June (697).
Same as Wright 77.

Ilicace^

Ilex Krugiana Loes. ; Engler's, Bot. Jahrb. 15, 317. Conch
Sound, May. Upper surface of the leaves shining, apex acute
(553)- Same as Cooper 17 from New Providence, and Eggers
1889 from St. Domingo (unnamed).

Sapindace^

Serjania diversifolia Radlk. "Fowl-foot vine." Common.
Conch Sound, May; Fresh Creek, Deep Creek, June (578, 687).
Same as Wright 109.

166 FLORA OF NEW PROVIDENCE AND ANDROS

Serjania subdentata Juss. & Poir. "Fowl-foot vine." Com-
mon in the coppet. New Providence, Jan.; Fresh Creek, June
(236). Same as Wright 2162.

Cardiospermum Halicacabum L. "Balloon vine." Nassau,
Jan. (237).

Thouinia discolor Griseb. "Quicksilver bush." Nicol's
Town, March; Conch Sound, May; Deep Creek, June (368,590,
686).

ExoTHEA PANicuLATA (Juss.) Radlk. Nicol's Town, March
(392). Same as Wright 1169.

Hypelate trifoliata Sw. "Ebony." Deep Creek, June (690).
Same as Wright 2 171.

Alvaradoa amorphioides Liebm. Nassau, Jan. (145). Same
as Wright 2189.

RHAMNACEiE

Krugiodendron ferreum (Vahl) Urban. (Condalia.) Fresh
Creek, June (611).

Reynosia Northropiana Urban, Symbolae Antillanae 3: 315.
1902.

Plate 10. Reynosia Northropiana. Portion of plant x f ; g, flower; m, interior
of calyx showing stamens and petals; a, stamen and petal; n, petal; d, pistil.

Red Bays on the west side of Andros, April 23, and at Nicol's
Town on the east side, April 28 (510).

Reynosia latifolia Griseb. Common on border of swash on
west side of Andros. Purser Point, June (661).

CoLUBRiNA FERRUGiNOSA Brongn. "Bittcrs." Deep Creek,
June (684). Same as Wright 1139.

GouANiA DoMiNGENSis L. " Chcw-stick." Common on both
islands. Nassau, Jan. (69).

VitacetE

ViTis ROTUNDiFOLiA Michx. Common. Conch Sound, May

(559)-

Cissus MiCROCARPA Vahl. "Bull-vine." Common in the south-
ern part of Andros. Deep Creek, June (694). Same as Wright 72.

Cissus sicyoides L. Conch Sound, May (582). Same as
Wright 74.

Cissus, sp. "Bull-vine." Quite common in the northern part

Mem. Torrey Club, 12

Pl. 10

:^f A§/ IS

d^ ^

REYNOSIA NORTIIROPIANA URBAN

HELIOTYPE CO., BOSTON.

ToRREY Club, 12

Pl. I

HELICTERES SPIRALIS

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 167

of Andros. Conch Sound, March, May (398). Same as Wright

3514.

Parthenocissus quinquefolia (L.) Planch. {Ampelopsis
quinquefolia (L.) Michx.) Carmichael, New Providence, Feb.
(322). Single specimen.

TlLIACE^

Triumfetta semitriloba L. Common about Nassau. Jan.

(4, 47)-

CoRCHORUS HiRSUTUS L. Common on both islands. Nassau,
Jan.; Nicol's Town, March (202). Same as Wright 2091.

CoRCHORUS siLiQUOSUS L. Common about Nassau; very-
variable in the size of the leaves. Nassau, Jan. (139, 222).

Malvaceae

SiDA CARPiNiFOLiA L. Common. Nassau, Jan. (35). A very
strict form with crenate leaves found at Nicol's Town, April (433).

SiDA supiNA L'Her. Nassau, Feb. (306).

Abutilon crispum (L.) Medic. Nicol's Town, March (370).

Pavonia spicata Cav. (P. racemosa Sw.) Collected by Mr.
Alexander Keith at Conch Sound, 1891 (736).

Hibiscus cryptocarpus Rich. "Wild cotton." Nicol's Town,
March; Deep Creek, June (397). Same as Wright 1575.

Hibiscus esculentus L. "Gumbo," "okra." Conch Sound,
May. Cultivated and escaped (549).

Hibiscus tiliaceus L. Not uncommon along the east coast
of Andros. Conch Sound, May (564). Same as Eggers 2632.

Ceiba pentandra (L.) Gaertn. {Eriodendron anfractuosum
DC.) "Silk-cotton tree," "Ceiba." Nassau, cultivated and es-
caped? Nassau, Jan. (152).

Sterculiace^
Helicteres semitriloba Bertero. Fresh Creek, June (664).

Helicteres spiralis sp. no v.

A tall shrub; young branches and leaves tawny tomentose;
leaves lance-ovate, acute, unequal and cordate at base, 6-9.5 c^-
long, 2.5-4 cm. wide, 3-5-ribbed, margin crenulate-denticulate,
sometimes with a few large teeth, upper surface pubescent, under

168 FLORA OF NEW PROVIDENCE AND ANDROS

tomentose, hairs tawny, stellate; petiole about i cm. in length;
stipules subulate; inflorescence fascicled, few-flowered, peduncles
1.5-2 cm. long, pedicels glandular at base; calyx campanulate,
1.5-2 cm. long, slightly bilabiate, unequally 5-toothed, lobes acute,
densely tomentose, about one- fourth the length of the column ; petals
5, white, shortly exceeding the calyx, oblong, clawed, subequal, the
lower auricled near the base; column 7-8 cm. in length, declined,
covered with long mostly simple hairs ; stamens 10, arranged in pairs,
anthers divergent-oblong, staminodia 5, ligulate; ovary tomentose,
5-celled ; styles united, thickened at the apex ; pod twisted, oblong,
about 4.5 cm. in length; convolutions about 10, tomentose or at
length glabrous.

Collected near the mouth of Fresh Creek, Andros, June; also
collected at Conch Sound by Mr. Alexander Keith, 1891 (645).

Plate ii. Helicteres spiralis. Portions of plant in flower and fruit X ^ nearly.

Melochia nodiflora Sw. Nassau, Jan. (45). Same as Wright

39-

Melochia tomentosa L. Common on both islands. Nassau,

Jan.; Nicol's Town, March (253). Same as Wright 40; same as

Eggers 1991.

Waltheria Americana L. Common on both islands; very
variable. Nassau, Jan.; Nicol's Town, March (89, 230, 429).

W. Americana var. ? (136, 207). Collected at two different
stations in Nassau, differing from all specimens examined in hav-
ing leaves smaller, 1-2 cm. long, 7-13 mm. wide, marginal teeth
not as acute, and in having both leaves and stems covered with
stellate hairs ; plant tall, less stout than the type.

HYPERICACE.E

AscYRUM HYPERicoiDES L. Common in the pines on both
islands. Nassau, Jan.; Conch Sound, May (323). Same as Wright
2129 and Eggers 2047 (unnamed).

BlXACEiE

Xylosma ilicifolia sp. no v.

Shrub 2-3 m. in height, bark grayish, verrucose, usually armed
with slender spines 1-4 cm. long, spines often much branched ; leaves
alternate, variable in shape, oval, oblanceolate or obovate, 2-3 cm.
in length, 1-2 cm. wide, coriaceous, shining above, apex acute and

[em. Torre y Ciatb, 12

XYLOSMA ILICIPOLIA

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 169

strongly mucronate (or sometimes obtuse), margin entire or frequently
with one to three large, mucronate teeth on the upper half of the leaf,
base cuneate or sometimes obtuse, petiole 1-2 mm. in length; flowers
dioecious, minute; staminate greenish, in axillary fascicles of 5-6,
very short-pediceled, bracts minute, ciliate; sepals 4-5, ovate,
ciliate at apex; stamens 8-10, filaments recurved, longer than the
anthers, anthers globose, disk annular; pistillate flowers 2-3 in a
cluster, pedicels about i mm. in length; sepals 4, lanceolate, ciliate
towards the apex, pistil more than twice the length of the sepals,
about I mm. long, ovary globose, styles two, thick, short, stigmas
two, disk annular, berry globose, bluish black, about 5 mm. in
diameter, one-celled and four-seeded (124, 388).

Collected at Nassau, Jan. ; Nicol's Town, Andros, March 27
and April 9; Fresh Creek, June 10. It is the same as Cooper's
13 from New Providence, marked in pencil by Gray " Xylosma
infestum?^^ ; also collected by Governor Robinson 322; Brace 98 and
Eggers 446. Seems to be between X. infestum Griseb. and X. huxi-
folium A. Gray ; differs from the latter in having the upper surface
of the leaves shining, apex mucronate, sepals ciliate, petioles and
pedicels shorter, and in the shape and margin of the leaves.

Plate 12. Xylosma ilicifolia. Portion of plant, X f; n, fruit; &&, flower;
a, sepal; c, stamens; e, ovary.

Canellace^

Canella WiNTERANA (L.) Gacrtn. (C. a/6d^ Murray.) "Wild
cinnamon," "Bahama whitewood bark." Not uncommon. Nassau,
Jan.; Deep Creek, June (79, 708). Same as Wright 2122.

TURNERACE^

TuRNERA ULMiFOLiA L. Common on both islands. Nassau,
Jan. (57). Same as Wright 209.

Passiflorace^

Passiflora angustifolia Sw. Nicol's Town, Conch Sound,
March (389, 427).

Passiflora cuprea L. "Wild watermelon." Common on
both islands. Nassau, Salt Cay, Jan.; Nicol's Town, March (121,

243)-

Passiflora minima L. Common. Nassau, Jan. ; Salt Cay,
Conch Sound, March (216, 242, 428). Same as Wright 1245.

170 FLORA OF NEW PROVIDENCE AND ANDROS

Passiflora multiflora L. NicoPs Town, March (374). De-
termined at Kew by Dr. Masters.

Passiflora pectinata Griseb. Common on Andros in the
pines. Nicol's Town, Conch Sound, March; Red Bays, April

(391)-

Determined at Kew by Dr. Masters, who noted, "I do not like
to, separate this from P. pectinata, though it differs." The following
constant characteristics were found in all the specimens collected.
Leaves deeply contracted below the middle with marginal stipitate
glands ; petiole with a few stiputate glands towards apex ; peduncles
as long or longer than the leaves ; leaflets of involucre bipinnatipartite,
secondary segments long-setaceous; excurrent in a gland; calyx
segments ovate-lanceolate; petals oblong; tendrils longer than the
leaves.

Caricace^

Carica Papaya L. "Papaw." Deep Creek, July. Culti-
vated (132).

Cactace^

Cereus Swartzii Griseb. "Dildo." On borders of creeks in
the southern part of Andros. Deep Creek, June (699). 4 mm.
in height, 7.5-10 cm. in diameter.

Opuntia spinosissima Mill. Along the shore. Fresh Creek,
June (652).

Opuntia Tuna (L.) Mill. Along the shore. Fresh Creek,
June (633).

Lythrace^

Parsonsia Parsonsia (L.) Britton. (Cuphea Parsonsia (L.)
R. Br.) Near Southwest Beach, N.P., Feb. (318). Same as
Eggers 1657.

Rhizophorace^

Rhizophora Mangle L. "Mangrove." Common on both
islands; the chief vegetation of the swash. Nassau, Jan. (193).

Myrtace^

Calyptranthes pallens Griseb. "White stopper." Com-
mon in the coppet. Fresh Creek, Dssp Creek, June (641).

FLORA OF NEW PROVIDENCE AND ANDROS 171

Myrtus punctata Griseb. {Anamomis.) "Naked wood."
Deep Creek, June (696).

Eugenia axillaris (Sw.) Willd. ''Stopper." Deep Creek,
June (707).

Eugenia longpies Berg. Red Bays, April; Conch Sound,
May (471). Same as Curtiss 985 from Florida.

Eugenia monticola DC. Deep Creek, July. In fruit only
(725). Same as Curtiss 987.

Eugenia sp. ? Near E. longipes Berg.

A tree with smooth, white bark, branchlets slender, smooth;
leaves thick, at length leathery, pale green, shining above with
pellucid dots; oblanceolate or elliptical, obtuse, mostly cuneate at
base, 10-17 ^^- long, 5-7 mm. wide, petioles 1-2 mm., veining
indistinct; flowers solitary or in pairs on long slender pedicels, 2-
2.5 cm. long, bibracteolate, bractlets oblong, about equaling the
calyx lobes. Flowers smaller than in E. longipes. In general
appearance the tree resembles Myrtus punctata except that the leaves
are smaller.

A single tree seen in the coppet at Deep Creek, July (722).

PsiDiUM GuAYAVA L. " Guava." Common in cultivation.
Nassau, Conch Sound (579).

Combretace^

CoNOCARPUS ERECTA L. " Button-wood." Abundant on both
islands. Nassau, Jan.; Cocoanut Point, May (53, 298).

CoNOCARPUS ERECTA SERiCEA Fors. Nassau, Cocoanut Point
(81, 532). Same as Eggers 2608. At Cocoanut Point, both the
glabrous type and the variety were growing together with no inter-
mediate forms.

Laguncularia racemosa (L.) Gaertn. "Bastard button-wood."
Common, especially in the swash. Fresh Creek, June (594).

Terminalia spinosa sp. nov. "Brier tree," "Prickly tree"

A low, spreading, flat -topped tree, height about 5 meters, 1.5-
2 dm. in diameter, bark lightish, ridged; branches horizontal,
branchlets divaricate, spiny; leaves fascicled, subsessile, oblanceo-
late or spatulate, 1-1.5 cm. long, 4-6 mm. wide, thick, glabrous,
yellowish green below, obtuse or retuse at the apex, tapering into

172 FLORA OF NEW PROVIDENCE AND ANDROS

a short petiole at base (1-2 mm.), margin entire, slightly revolute;
spines axillary, slender, 3-7 mm. in length, mostly in threes at the
end of the branchlets; flowers small, greenish, growing in axillary
spikes, the 5-7 flowers approximate, so as to resemble a head; pe-
duncle 6-10 mm., bracts small, ovate; calyx valvate, tube 1-2.5 ^^*
long, prolonged beyond the ovary, constricted above it, limb campan-
ulate, subtruncate, 1-2 mm. long, with 5 small teeth, villous within,
deciduous, disk of 4 brownish, villous, two-lobed glands at mouth of
calyx tube; stamens 8 (9), exserted, about 3 mm. in length, alternate
ones inserted lower down on the calyx, filaments slender; anthers
cordate; ovary i-celled; ovules 3, flask-shaped or oblong, suspended
from the top of the cell; style simple, equaling or slightly exceeding
the stamens, subulate, a little thickened at base, villous; stigma
simple, young fruit ovoid. Foliage closely resembling that of T.
angustifolia but inflorescence very different.

A few monstrous flowers were found like those described by
Grisebach as occurring in Bucida Buceras L. Collected on the sa-
vannas at Red Bays in April and near the source of Fresh Creek
in June, no. (502).

Plate 13. Terminalia spinosa. Portion of plant, natural size; n, cluster of
flowers; a, flowers showing disk; p, stamen.

Buceras Catappa (L.) A. S. Hitch. "Almond tree." Nassau,
Jan. Cultivated.

Melastomace^

Tetrazygia bicolor (Mill.) Cogn. {T. elceagnoides DC.) Com-
mon in the pines on both islands. Nassau, Jan.; Conch Sound,
May (127). Same as Wright 1222.
>

Onagrace^

Jussi^A suFFRUTicosA L. Near Southwest Beach, N.P., Feb.
(320). Same as Wright 159.

Samydace^

Casearia l^tioides (Rich.) Cocoanut Point, Andros, April
(514). Same as Wright 1108.

Casearia Bahamensis Urban. Nicol's Town, March, April
(384). Distributed as Thioidia serrata Endl.

Banara reticulata Oris. Conch Sound, May (558). Same
as Wright 1882. Determined at Kew by Dr. N. L. Britton.

Mem. Torrey Club, 12

Pl. 13

TERMINALIA SPINOSA

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 173

Umbellifer^

Hydrocotyle pygm^a Wright. Red Bays, Conch Sound,
April (499, 524). Determined at Gray herbarium by Mr. J. M.
Greenman.

Centella Asiatica (L.) Urban. {Hydrocotyle Asiatica L.) In
low ground in the pines. Red Bays, April (494).

Anethum graveolens L. "Dill-seed." Nicol's Town, April.
Escaped (438).

MYRSINACEiE

Rapania Guyanensis Aubl. {M. Floridana A. DC.) Nicol's
Town, March (387).

IcACOREA PANicuLATA (Nutt.) Sudw. (Ardisia Pickeringia T.
& G.) "Stopper-tree." New Providence, Jan. (234). Same as
Curtiss 1799.

Jacquinia Keyensis Mez. "Joe-bush," "iron wood." Com-
mon on the cays and along the eastern shores of both islands. Rose
Island, N.P., Feb.; Fresh Creek, June; Mars Bay, July (251).
A new species described in Symbolae Antillanae, 1890.

Primulace^

Samolus ebracteatus Kunth. Common on the savannas at
Red Bays, April (478).

Plumbaginace^

Plumbago scandens L. Common about Nassau, Jan. (11).

Sapotace^

Chrysophyllum oliviforme Lam. "Saffron-tree." In the
pines on both islands. New Providence, Feb. (262).

BuMELiA CuBENSis Griscb. London Creek, Andros, May. In
fruit only. "Rather small-leaved." Determined by Dr. B. L.
Robinson, Gray herbarium (544).

BuMELiA MiCROPHYLLA Griscb. In the swash on the west side
of Andros. In fruit only. Purser Point, June (666). Same as
Brace 234 and Eggers 4418. Determined by Dr. Britton at Kew.

LucuMA PAUCiFLORA A. DC. "Egg fruit." Deep Creek, June
(703). Same as Wright 346.

174 FLORA OF NEW PROVIDENCE AND ANDROS

SiDEROYXLON MASTiCHODENDRON Jacq. In fruit. Cocoanut
Point, Andros, April (450). Same as Wright 1324.

DiPHOLis SALiciroLiA A. DC. "Wild cassada," "bustic."
Common in the coppet on both islands. Nassau, Feb.; Fresh
Creek, June; Deep Creek, July (326). Same as Eggers 4106.

MiMUSOPS DissECTA R. Br. "Wild sapodilla." Common near
the shore on both islands. Hog Island, N.P., Feb. (263).

MiMUSOPS Floridana Engelm. "Wild sapodilla." Conch
Sound, May; Deep Creek, July (734). Same as Curtiss 1766,
Determined at Kew by Dr. N. L. Britton.

AcHRAS Sapota L. "Sapodilla." Common in cultivation.
Nassau (42).

Ebenace^

Maba Carib^a (A. DC.) Hiern. {Macreightia.) Fresh Creek,
June (640). Same as Wright 133 1.

OLEACEiE

Adelia porulosa (Poir.) Engler. Savannas back of Red Bays.
April (511). Determined by Dr. B. L. Robinson, Gray herbarium.

Adelia sp. Coppet, Conch Sound, May. Staminate flowers
only. Near A. porulosa^ but leaves are broader, 1.5-2 cm. wide,
5 cm. long, more tapering at the base and apex; veins inconspicuous,
not shining above; filaments broader (589).

Loganiace^

Spigelia Anthelmia L. Mastic Point,. May (599). Same as
Wright 390.

Cynoctonum Mitreola (L.) Britton. (Mitreola petiolata T. &
G.) In savannas on the west side of Andros. Differs from all speci-
mens examined in having the branches of the cymes much more
divergent, and the inflorescence scarcely, if at all, unilateral. Red
Bays, April (492).

Cynoctonum sessilifolia (T. & G.) Britton. 12-15 cm. in
height, strict, leaves rather rigid, 0.5-1 cm. in length, like specimen
from East Florida collected by Palmer, no. 436; also like speci-
men from Key West (573). In savannas on west side of Andros,
Red Bays, April.

XORA OF NEW PROVIDENCE AND ANDROS 175

Gentianace^

EusTOMA EXALTATUM Griscb. Abundant in the savannas on the
west side of Andros; occurs sparingly elsewhere. New Provi-
dence, Jan. ; Red Bays, April (201, 456).

Sabbatia campanulata (L.) Torr. {S. gracilis Salisb.) Quite
common on both islands. On New Providence growing in sand near
the shore; on Andros on the savannas near Red Bays and also near
fresh water in the interior. Very variable, flowers often white.
Hog Island, N.P., Feb.; Red Bays, April; Stafford Creek, May
(322, 464, 744).

VoYRiA Mexicana Griseb. Conch Sound, March (415).

APOCYNACEiE

ViNCA ROSEA L. Nassau, Jan. (in).

Plumiera obtusa L. "Frangipani," ''milkweed." Not un-
common near the shore. Fresh Creek, June (651).

EcHiTES Andrewsii Chapm. "Wild unction." Common.
Conch Sound, April, May (522).

EcHiTES BiFLORA Jacq. Conch Sound. Collected by Mr.
Alexander Keith, 1892 (740). Same as Eggers 2676.

EcHiTES Sagr^i a. DC. Common. Nassau, Jan.; Conch
Sound, March; Red Bays, April (128, 431, 461).

EcHiTES umbellata Jacq. "Devil's potato-root," "dream-
vine." Very common and variable; leaves range from lanceolate
to orbicular, 2-10 cm. in width. Nassau, Jan.; Conch Sound,
March; Red Bays, April; Deep Creek, June (61, 403, 673).

Plumiera rubra L. "Frangipani," "jessamine tree." Mastic
Point (601). _^

Thevetia Thevetia (L.) Millsp. Cultivated at Nassau (75).

ASCLEPIADACE^

AscLEPiAS CuRASSAViCA L. Common about Nassau, Jan. (21).
Metastelma Bahamense Griseb. Conch Sound, March (410).

Metastehna barbatum sp. no v.

Stem smooth, twining; leaves linear or linear-lanceolate, occa-
sionally oblong, 1-2.5 cm. in length, 2-3 mm. in width, apex cuspidate

176 FLORA OF NEW PROVIDENCE AND ANDROS

or obtuse and mucronate, occasionally acute, base obtuse, margin
entire, slightly thickened, midrib prominulous on the under side,
petioles 2-3 mm., umbels five- to eight-flowered, peduncle 1-1.5
mm., pedicels about i mm.; calyx lobes smooth, ovate, obtuse,
about one-fourth the length of the corolla; corolla greenish white,
urceolate-campanulate ; petals lanceolate-oblong, obtuse, united
about one-third of the way, the upper third densely villous within
and with a few scattered hairs in the center of the petal below,
crown segments attached to the base of gynostegium and of
the corolla, equaling the column in length, somewhat incurved,
ligulate, bifid at the apex, teeth often unequal, rarely obtuse ; gynos
tegium about half as long as the corolla; stigma depressed; follicle
slender, acuminate, 3 cm. in length (474).

Plate 14. Metastelma barbata. Portion of plant; e, crown; a, calyx and pistil;
c, petals; hy part of gynostegium; ^, pollinia.

A low, twining plant, common on the savannas near Red Bays
on the west side of Andros, April 15. The same as a specimen at
Kew collected by Governor Robinson in the pine barrens of New
Providence, April, 1877.

The species is closely related to Metastelma Blodgettii Gray, but
differs in having no lines of pubescence on the stem, umbels 5-8-
flowered, peduncles and pedicels shorter and the flowers smaller
with the calyx shorter in proportion to the corolla, scales of the
crown slightly incurved, broader and toothed at the apex.

Metastelma palustre Schltr. (Seutera maritima Decsn.)
Hog Island, N.P., Feb.; Red Bays, April (333, 475).

CONVOLVULACE^

Ipomcea Batatas Lam. "Sweet potato." Cultivated and
escaped on both islands. Nassau, Jan. (77, 223).

Ipomcea cathartica Poir. Nassau, Jan. (220). Determined
by Dr. Britton at Kew.

Ipomcea coccinea L. Nassau, Jan. (120).

Ipomcea commutata R. & S. Nassau, Jan. (231).

Ipomcea fastigiata Sweet. Nicol's Town, April (518). Com-
pared at Kew by Dr. Britton. Same as Eggers 4370, 4541 from
Bahama.

Ipomcea heptaphylla Griseb. In the pines. Conch Sound,
May (569). Same as Wright 1371.

Mem. Torrey Club, 12

Pl.H

METASTELMA BARBATA

HELIOTYPE CO., BOSTON,

FLORA OF NEW PROVIDENCE AND ANDROS 177

Ipomcea Jamaicensis Don. "Glory-morning." Common and
very variable. Nassau, Feb.; Conch Sound, April; Fresh Creek,
June; Mars Bay, July (77, 254, 451, 623, 710). Determined by
Dr. Britton at Kew.

Ipomcea grandiflora Lam. (/. longiflora R. Br.) On sandy
beaches. Salt Cay, N.P., Jan.; Deep Creek, June (244).

Ipomcea Pes-capr^ (L.) Sweet. ''Bay-hop." Very common
on sandy shores on both islands. Cocoanut Point, Andros, May

(533)-

Ipomcea repanda Jacq. (/. arenaria Steud.) Common in the
pine-yard. Conch Sound, March (394, 404). Same as Wright
3102. No. 394 has the leaves deeply cordate at base.

Ipomoea sidifolia Chois. "Christmas gambol," "Christmas-
flower." Common about Nassau. Nassau, Jan. (13).

Ipomcea sinuata Ort. (/. dissecta Pursh.) Common in the
coppet on both islands. Nassau, Jan.; Fresh Creek, June (16, 76).

Ipomoea triloba L. Conch Sound, March; Fresh Creek, June.
Leaves 1.5-2 cm. long, plant smaller and more delicate than any
specimens examined; closely resembles an unnamed specimen of
Blodgett's from Key West, in Torrey herb (423).

Jacquemontia Jamaicensis (Jacq.) Hall. Common on both
islands. Nassau, Jan. (135).

Jacquemontia verticillata (L.) Urban. Conch Sound, March
(424). Same as Wright 455.

EvoLVULUs ARBUSCULA Poir. Frcsh Creek, June (607). Re-
sembles Wright 1658; differs from Wright 456 in being more slen-
der, not as erect, flowers smaller, 3-4 mm., white; calyx and corolla
sericeous; calyx shorter, about 1.5 mm. long. Plant about 6 dm.
in height, leaves 1-2 mm. long.

DicHONDRA repens Forst. Nassau, Jan. (172). Same as
Wright 459.

CuSCUTACEtE

CuscuTA Americana L. Growing on shrubs in low ground.
Red Bays, April (497). Same as Wright 1659.

Hydrophyllace^
Nama Jamaicense L. Nassau, Jan. (166),

178 FLORA OF NEW PROVIDENCE AND ANDROS

BORAGINACE^

CoRDiA ANGUSTiFOLiA R. & S. Fresh Creek, June (619).
Agrees with Wright 3 114 except in having spatulate leaves and
surface much less pubescent.

CoRDiA LIMA R. & S. " Granny-bush." Common on Andros.
Nicol's Town, March (376). Ex. desc.

CoRDiA Sebestena L. Common along shore on both islands.
Same as Cooper 52; agrees with Wright 3554 except in the shape of
the leaves. All the Bahaman specimens examined have the leaves
oval instead of ovate (107).

BouRRERiA Havanensis (Willd.) Miers. ^'Strong-back." Com-
mon on both islands in the pines and coppet. Nassau, Jan. ; London
Creek, May; Fresh Creek, June (74, 542).

TouRNEFORTiA GNAPHALODES (Jacq.) R. Br. Common on
sandy beaches on both islands. Nassau, Jan. (63).

TouRNEFORTiA voLUBiLis L. Nassau, Jan.; common on both
islands (212). .

Heliotropium Curassavicum L. Nassau, Jan. ; Middle Bight,
Andros, June (198).

Heliotropium parviflorum L. Nassau, Jan. (24).

Heliotropium nanimi sp. nov.

Low, shrubby, corymbose-branching, 8-1 1 cm. high, branches
erect-ascending, entire plant strigose-canescent ; leaves numerous,
appressed, alternate, sessile, about 2 mm. long and i mm. wide,
elliptical-oblong in shape, acute, margin entire; flowers scattered,
solitary, sessile, about as long as the leaves and opposite them;
calyx persistent, segments 5, imbricate, slightly unequal, lanceolate,
acute, about i mm. in length, strigose-canescent; corolla white,
campanulate, shortly exceeding the calyx, hairy on the outside,
lobes 5, ovate-oblong; stamens inserted about the middle of the corolla
tube, anthers lanceolate, appendiculate appendage almost as long
as the anther, disk fiat, ovary free, globose; style short; stigma
annular with a conical tip; fruit depressed-globose, separating into
four one-seeded, hairy nutlets; seeds curved (757).

Collected at Red Bays on the savannas. In flat, rather marly
ground near the shore.

FLORA OF NEW PROVIDENCE AND ANDROS 179

Heliotropium manum. a, Plant X | ; &, ovary ; c, calyx.

Verbenace^

Lantana Camara L. Not uncommon. Nassau, Jan.; Red
Bays, April (129, 490).

Lantana crocea Jacq. Nicol's Town, March; Conch Sound,
^3,y (352, 561). 352 has larger leaves, 3.5-4.5 cm. in length, and
shorter peduncles (1.5-2 cm.) and may be L. polyacantha Schauer,

180 FLORA OF NEW PROVIDENCE AND ANDROS

as it much resembles a specimen collected by Eggers at St. Thomas
and so named by Watson.

Lantana involucrata L. "Wild sage," "big sage." Com-
mon on both islands. Fort Montague, Nassau, Jan. (56).

LippiA CANESCENS Kth. Nassau, Jan. (114).

LiPPiA NODiFLORA (L.) Michx. Hog Island, N.P. ; Red Bays,
April (250, 481). Same as Eggers 1537, St. Domingo (unnamed).

Abena Jamaicensis (L.) A. S. Hitch. {Stachytarpha Vahl.)
Roadsides, Nassau, Jan. (19).

CiTHAREXYLUM Berterii Sprcng. Calabash Cay, Andros,
June (608). Leaves much longer than in Wright (1356), 10-12 cm.
long, 1-2.5 c^- wide.

CiTHAREXYLUM LUCiDUM Cham. & Schlccht. Leaves oblance-
olate, cuneate at base, shining and leathery when old, apex obtuse
or emarginate, corolla puberulous, raceme erect in fruit. Conch
Sound, May (571).

DuRANTA REPENS L. {D. PlumieH Jacq.) Common on both
islands. Nassau, Jan. (39).

Petitia Domingensis Jacq. Nicol's Town, March (358).
Compared at Kew by Dr. Britton. Same as Eggers 4203 and Robin-
son 55.

ViTEX iLiciFOLiA Rich. Fresh Creek, June (625). Deter-
mined at Kew by Dr. Britton. Same as Wright 3180.

AviCENNiA NiTiDA Jacq. "Salt bush," "black mangrove."
Common in the swash and along the shore. Mastic Point, May

(593)-

OviEDA FRAGRANS (W.) A. S. Hitch. (Clerodendrofi.) "Wild
jessamine." New Providence, along the roadside. Naturalized.
Feb. (328).

OviEDA ACULEATA (L.) A. S. Hitch. Nassau, March. Escaped?
(296).

Labiate

MiCROMERiA Brownei Benth. Conch Sound, April (526).
Petioles 4-5 mm., equaling or exceeding the leaves and peduncles:
under surface of leaves and stem purplish.

Salvia occidentalis Sw. New Providence, Feb. (265).

Salvia serotina L. Silver Cay, Nassau, Jan. (157).

FLORA OF NEW PROVIDENCE AND ANDROS 181

Leonurus Sibiricus L. Common about Nassau, Jan. (2).
Leonotis nepet^folia (L.) R. Br. Nassau, Jan. (31).
Teucrium Cubense L. Nassau, Jan. (108).
Mesosph^rum pectinatum (Poit.) Kuntze. Red Bays, Andros,
April (505).

SOLANACE^

SoLANUM ACULEATissiMUM Jacq. Nassau, Jan.; Fresh Creek,
June.

SoLANUM Bahamense L. '' Cankerbcrry." Common on both
islands. Variable. Nassau, Jan.; Salt Cay, Jan. (174, 241).
174 has prickles on both veins and midrib of the leaves as well as on
the stems while 241 is entirely unarmed and has racemes 10-12 cm.
long, recurved at the apex.

SoLANUM NIGRUM L. Nassau, Jan. (126, 228).

SoLANUM NIGRUM NODiFLORUM Gray. Conch Sound, May;
Fresh Creek, June (557, 614).

SoLANUM VERBASCiFOLiUM L. " Salvc-bush." Common on both
isla^ds. Nassau, Jan. (131).

Physalis angulata L. Fresh Creek, June (616).

Physalis Barbadensis Jacq. Conch Sound, March (421).

Capsicum baccatum L. "Bird-pepper." Hog Island, N.P.,
Feb. (256).

Oestrum pallidum Lam. Nicol's Town, April (432).

Datura Metel L. Deep Creek, June. Probably escaped
(700).

Scrophulariace^

Antirrhinum antirrhiniflora (Willd.) A. S. Hitch. {A,
maurandioides Gray.) Nassau, Jan. Probably escaped (95).

Russelia juncea Zucc. Escaped. Nassau, Jan. (122).

Stemodia maritima L. Nassau, Jan. (265).

Capraria biflora L. Common on both islands. Nicol's
Town, March (27, 381).

BucHNERA ELONGATA Sw. Common on savannas on the west
side of Andros. Red Bays, April (467).

Gerardia maritima Raf. Common on savannas on west side
of Andros. Red Bays, April; Purser Point, June. Variable (751).

182 FLORA OF NEW PROVIDENCE AND ANDROS

Gerardia purpurea L. 4-4.5 ^^' ^^ height, mostly simple
and strict. Red Bays, April (459).

Lentibulariace^

Utricularia foliosa L. Stafford Creek, Andros, May. ''No
certain determination possible without flowers : the bladders appear
like those of U. foliosa which is common in the West Indies." Dr.
Thomas Morong (543, 547).

Utricularia gibba L. Stafford Creek, Andros, May. No
flowers. "Leaves and bladders much resemble those of U. gibba
which occur in Florida." Dr. Thomas Morong (548).

PiNGUicuLA PUMiLA Michx. Rare; on the edge of the swash
on the west side of Andros. Red Bays, May (572).

BiGNONIACEiE

Tecoma lepidophylla Griseb. Purser Point, Andros, June
(660). Same as Wright 13 41.
Tecoma sp.

Medium-sized shrub, twigs rusty-lepidote ; leaves digitate, with
three or five leaflets, petioles 1-2 cm. long, channeled above, spar-
ingly rusty-lepidote, petiolules 2.5 mm. long, central, one-half as
long again as the lateral; leaflets 2.5-4.5 cm. long, 1.5-2.5 cm.
broad, obovate or oval, retuse or emarginate and mucronate at apex,
►obtuse and usually unequal at base, margin crenulate, veins promi-
nent below; leaves thick, sparingly lepidote above; scales abundant
b)eneath, the larger ones rusty, giving the under surface a slightly
brownish tint; flowers not seen; calyx bilabiate, rusty-lepidote, at
length deciduous, 7-10 mm. in length, lobes acute; capsule linear,
pointed, subcompressed, 6-9 cm. long, rusty-lepidote; valves slightly
keeled.

Collected at Long Ridge Cay, Andros, June 20 (758).

Tecoma Bahamensis sp. nov. "Beef-bush"

A tall shrub, twigs light brown, lepidote; leaves opposite, pal-
mately compound, with five leaflets, petioles 3-5 cm. in length,
flattened and channeled above; leaflets all petiolulate, petiolule
of the central leaflet about i cm. in length, being slightly longer
than those of the lateral leaflets and more than twice as long as those
of the basal leaflets, leaflets oblong or oblong-elliptical, sometimes
oblanceolate, 2-4 cm. long, 1-2.5 ^^- wide, the central leaflet largest,

Mem. Torrey Club, 12

TECOMA BAHAMENSIS

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 183

thickish, apex obtuse or retuse, margin entire or slightly undulate,
veins prominulous beneath, 4-7 pairs diverging from the midrib at
nearly right angles, surface lepidote on both sides, pale green above,
white beneath owing to the confluent white scales; inflorescence
terminal, sessile, many-flowered cymes; flowers showy, pale pink,
pedicels rusty-lepidote, about i cm. in length; calyx rusty, campanu-
late, closed in the bud, about 1.5 cm. in length, bilabiate, lower lip
subtruncate, upper longer, usually rounded; corolla funnel-shaped,
veiny, pink, about four times as long as the calyx, limb spreading,
lobes rounded, slightly unequal, margin sometimes undulate, tube
pubescent within ; stamens 4, didynamous, less than half the length
of the corolla, inserted near the base of the tube, filaments incurved,
anther cells linear-oblong, divaricate, straight; pistil about 2.5 cm.
in length; ovary lepidote, two-celled; style slender; stigma two-
lamellate, lamellae cuneate or obovate; disk pulvinate; capsule
elongated, linear, about 8 cm. in length and 7-8 mm. in width,
slightly compressed, rusty-lepidote, valves coriaceous, subcarinate,
veiny ; seeds oblong, slightly notched below, about as long as the striate
hyaline wing (218).

Collected on New Providence, near Nassau, Jan. 26; in the
interior of the island, Feb. 26, and at Nicol's Town, Andros, March 26.

Note from Kew, 1899: "This matches a plant collected in the
Bahamas by Brace no. 643 and Eggers no. 3962 which has been
called Tabebuia leucoxyla DC, but according to Bureau T. leucoxyla
is identical with T. obtusifolia Bureau, a totally different plant with
simple leaves." Also resembles a specimen at Cambridge from the
Bahamas 1859 marked T. leucoxyla ? by Gray.

Plate 15. Tecoma Bahamensis. Portion of plant in flower, X f.

Tecoma stans (L.) Juss. ''Yellow elder." Common about
Nassau, Jan. (8).

Jacaranda Bahamensis R. Br. "Boxwood," "cancer tree."
Deep Creek, June (701). Determined at Kew by Dr. Britton.

Jacaranda ccerulea Griseb. Nassau, Jan. (213). Same as
Eggers 4427.

ACANTHACEJE

Thunbergia fragrans Roxb. Probably escaped. Nassau,
Jan. (130).

Blechum Brownei Juss. Nassau, Jan. (49).

Anth ACANTHUS AcicuLARis (Sw.) Nces. Common in the coppet

184 FLORA OF NEW PROVIDENCE AND ANDROS

on Andros. Nicol's Town, April; Fresh Creek, June (396). The
same as a specimen of Cooper's from New Providence.
DiCLiPTERA ASSURGENS Juss. Nassau, Jan. (72).

RUBIACE^

ExosTEMMA Carib^um (Jacq.) R. & S. "Princewood." Kemp
Sound, Andros, June (685). Same as Wright 2674 and Curtiss 1132.

Rhacicallis Americana (Jacq.) A. S. Hitch. (R. rupestris
DC.) ''Sand-fly bush," "salt-water bush," ''seaweed," "wild
thyme." Common along the rocky shores of both islands. Red
Bays, April; Fresh Creek, June (458). Same as Wright 2696.

Hamelia patens Jacq. Nassau, Jan. (40). Same as Curtiss
5500 from Florida.

Catesb^a spinosa L. "Prickly apple." Fresh Creek, June
(624).

Catesbaea fasciculata sp. no v.

A low, spiny shrub with long, slender branches; leaves fascicled,
obovate or oblanceolate, 5.7 mm. in length, 3-4 mm. in width, ob-
tuse, tapering into a short petiole at base, thick, shining a'bove,
margin entire or slightly revolute, surface of the stem, spines and
upper surface of leaves minutely papillose; spines axillary, slender,
in pairs, about as long as the leaves; flowers scattered, small, soli-
tary, sessile in the axils; calyx-tube short, ovoid, lobes 4, subulate,
persistent; corolla valvate, campanulate, white, 5-7 mm. in length,
lobes 4, short, 1-2 mm. long, obtuse, spreading; stamens 4, inserted
at the base of the corolla tube, filaments glabrous, slightly exceeding
the tube, anthers linear, ovary 2-celled, 2-flattened, pendulous
ovules in each cell ; style smooth, a little longer than the stamens ;
fruit (immature) a berry.

Collected at Fresh Creek, Andros, June 6. The same as Eggers
4508 from Hog Island (627).

Plate 16. Cateshcea fasciculata. Portion of plant, about natural size; 0, flower
with corolla removed; w, interior of corolla showing stamens.

SCOLOSANTHUS Sp.

A low, tortuous branching shrub, slightly resinous, with slender
scattered spines, 5-6 mm. in length; leaves and stem minutely
papillose; leaves fascicled or opposite, minute, 2-2.5 ^^- ^^ length,
short-petioled, ovate, obtuse with revolute margins, thick, shining;
stipules small, connate; flowers not seen; drupe ovoid or globose,

CM. lORREY CLUB, 12

CATESBAEA FASCICULATA

4ELI0TYPE CO., BOSTOh

f FLORA OF NEW PROVIDENCE AND ANDROS 185

sessile, axillary, white, about 2 mm. in length, and containing one com-
pressed seed.

A single specimen collected on the south side of Fresh Creek,
Andros, June (646).

Randia aculeata L. Common on both islands. Nicol's
Town, March (383). Same as Wright 392 and Curtiss 11 29.

Genipa clusi^folia (Jacq.) Griseb. "Seven-year apple,"
Spruce Cay, N.P., Feb.; Mastic Point, Fresh Creek, June (299).
Same as Wright 3574 and Curtiss 1130.

GuETTARDA ELLIPTICA Sw. Stafford Creek, Andros, May;
Lisbon Creek, June (540, 677).

GuETTARDA SCABRA Lam. Stafford Creek, May; Fresh Creek,
June; Conch Sound, July (535, 649, 730). The specimens from
Stafford and Fresh Creek 535 and 649 are identical but differ greatly
from 730; the former have thicker, rigid leaves, paler in color, with
an entire revolute margin and the veins prominently raised on the
under surface. No. 730 has the leaves strongly mucronate, margin
crenate and the upper surface much more scabrous than in Wright
2707. No. 730 is the same as Brace 186 and 197 as compared at
Kew by Dr. Britton.

Antirrhcea lucida Gaertn. Deep Creek, July (724). Same
as Wright 1270.

Antirrhcea myrtifolia Griseb. Red Bays, April; Fresh
Creek, June (470). Same as Wright 2782 and Brace 445. Deter-
mined at Kew by Dr. Britton.

Erithalis fruticosa L. ''Black torch." Common on both
islands. Nicol's Town, March; Red Bays, April; Deep Creek,
June (365, 482, 691). Same as Curtiss 11 27.

Erithalis rotund ata Griseb. Deep Creek, June (739).
Same as Wright 1268.

Chiococca parvifolia Wulschl. ''Snake-root." Common in
the coppet on both islands. Nassau, Jan. ; Deep Creek, June
(138, 6SS). Same as Wright 3584.

Chiococca alba (L.) A. S. Hitch. (C. racemosa Jacq.) Nassau,
Jan. (28).

Chiococca sp. Red Bays, April (477).

The latter was a depressed form found in the savannas on the
west side of Andros. Stems erect, unbranched ; plant low, about 3

186 FLORA OF NEW PROVIDENCE AND ANDROS

dm. in height, with smaller, thicker, more rigid leaves, lanceolate with
a blunt apex, 2-2.5 c^- ^^^gj flowers yellowish brown. Resembles
a specimen of Blodgett's from Key West in the Gray herbarium.

Phialanthus myrtilloides Griseb. "Candlewood." Staf-
ford Creek, May; Fresh Creek, June (541, 642, 728).

Strumpfia maritima Jacq. Common on the cays and along
the shores of both islands. Nassau, Jan.; Lisbon Creek, Andros,
June (151).

MoRiNDA RoYOC L. "Wild mulberry," "wild pineapple." Not
uncommon in the pines on Andros. Conch Sound, March (425).
Same as Wright 2757. Plants 1.5-2 m. in height.

Myrstiphyllum pubescens (Sw.) a. S. Hitch. (Psychotria.)
Conch Sound, May (585). Same as Wright 243 and 1278.

Myrstiphyllum undatum (Jacq.) A. S. Hitch. {Psychotria.)
"Wild coffee." Nicol's Town, March (361).

Myrstiphyllum ligustrifolium sp. nov.

A low shrub with smooth dark bark; branchlets slightly angled;
leaves opposite, petioles 2-4 mm. in length, blade elliptical or
oblanceolate, 3-6 cm. long, 1-2 cm. broad, thickish, paler beneath,
apex acuminate, base cuneate or tapering, margin entire, slightly
revolute, glabrous or with a few scattered hairs beneath on the midrib
or in the axils of the veins, veins rather inconspicuous above ; stipules
rusty membranaceous, truncate, sheath deciduous, about 5 mm. in
length; flowers in three- to five-rayed terminal panicles, peduncle
about 2 cm. in length, pedicels about i mm. or flowers sessile, bracts
minute, ciliate; calyx deciduous, tubular-campanulate, 1-2 mm.
long, lobes 5, very short, acute, ciliate; corolla whitish, somewhat
funnel-form, about three times as long as the calyx, throat bearded,
lobes 5, valvate, half as long as the tube, lance-oblong, obtuse, thick-
ened and involute at the apex, at length reflexed; stamens 5 (6),
inserted in the throat of the corolla, included, filaments short, anthers
oblong; ovary two-celled, globose; style smooth, dilated upwards^
two-cleft at apex; drupe dark red, globose or ovoid, 5-6 mm. high,
4-5 mm. broad, pyrenae flat, four-furrowed, crests broadly obtuse (206).

Collected at two places in the neighborhood of Nassau, Jan.
The same as Eggers 4052 from Hog Island. It is related to M.
celastroides Gris.

Plate 17. Myrstiphyllum ligustrifolium. Branch in leaf and flower, about natu-
ral size; h^ fruiting branch; 5, single flower; a, interior of corolla showing stamens;
c, pistil.

Mem. Torrey Club, 12

Pl. 17

MYRSTIPHYLLUM LIGUSTRIFOLIUM

HELIOTYPE CO., BOSTON.

Mem. Torrey Club, 12

Pl. 18

A'r

ANGURIA KEITHII

HELIOTYPE CO., BOSTO^

FLORA OF NEW PROVIDENCE AND ANDROS 187

Ernodea littoralis Sw. Common in the pines on both islands
(sometimes near the shore). Flowers red or white. New Provi-
dence, Jan.; Rose Island, N.P., Feb.; Nicol's Town, March (102,
264).

Spermacoce tenuior Lam. In the pines, N.P., Feb. (319).
Same as Eggers 4441.

Galium hispidulum Mx. Conch Sound, April (523). Plants
glabrous, in leaf only.

CuCURBITACEiE

Anguria Keithii sp. nov.

Stem climbing, glabrescent, somewhat sulcate; leaves deeply
divided with seven narrowly elliptical segments, 4-6.5 cm. long,
.5-1 cm. wide, outside segments lobed near the base; central segments
the narrowest, apex mucronate, tapering at base, margin entire
above but the three or five central segments with a few large mu-
cronate teeth (one to five on each side); petiole .5-2 cm.; leaves
rather thin, somewhat punctate beneath; tendrils simple, much
longer than the leaves; inflorescence racemose, the 3-6 flowers
approximate at the top of the peduncle; peduncle about 8 cm.
long, longer than the leaves, pedicels 5-9 mm. long; staminate
flowers, calyx tubular-campanulate, constricted at the throat, tube
5-7 mm. long, lobes 5, 2-3 mm. long, lanceolate, acuminate; petals
orange, elliptical, about i cm. in length, obtuse or shortly mucronate ;
stamens 2, included, anthers inserted about the middle of the tube,
sessile, lanceolate, acuminate or acute, about 6 mm. in length, repli-
cate below about one-third of the way ; pistillate flowers and fruit not
seen (556).

Collected at Conch Sound, Andros, May 8. Comes nearest to
A. pedata Jacq., but differs in having leaves 7-cleft, segments narrower,
margin more deeply toothed, and leaves thinner ; flowers fewer, sepals
longer and acuminate.

Named for Mr. Alexander Keith, of Edinburgh, on whose sisal
plantation the plant was collected.

Plate 18. Anguria Keithii. Portion of vine, X f; a, stamen, front view; n,
stamen, rear view.

Trianosperma racemosa (Griseb.). T. & G. Conch Sound,
March (419). Same as Wright 1243.

188 FLORA OF NEW PROVIDENCE AND ANDROS

GOODENIACE^

Sc^voLA Plumieri (L.) Vahl. Common on the sandy beaches
on both islands. South Beach, N.P., Jan. (85).

COMPOSIT.E

Vernonia Bahamensis Griseb. Common in the pines on both
islands. Nassau, Jan. (loi). Same as Eggers 4187 and Brace 118.

Ageratum conyzoides L. Nassau, Jan. (22).

EuPATORiUM AGERATiroLiUM DC. Nassau, Jan. (176). Same
as Wright 2803.

Eupatorium Bahamense sp. no v.

Shrubby, branching, branches cylindrical, striate, tips pubes-
cent, somewhat rusty; leaves opposite, petioles 3-10 mm. in length,
lanceolate or ovate-lanceolate, 2-4.5 c^- lc)^gj •1S~^ cm. wide, base
cuneate, apex obtuse, margin entire or slightly repand, triply nerved
(in some young leaves obscurely so), the lateral nerves usually start-
ing 2-5 mm. above the base, glabrous above, glabrate and densely
dotted with glands beneath, glands mostly black ; corymbs numerous,
trichotomous, peduncles with a few scattered glands ; heads shortly
pedicellate, often in pairs, pedicels 2-6 mm. in length, ten- to thirteen-
fiowered; flowers blue, receptacle cylindrical, flat on top; involucre
cylindrical, 8-10 mm. long, about 2 mm. broad, scales imbricate in
about four rows, innermost linear-lanceolate, about 6 mm. in length,
outer lanceolate or oblong-lanceolate, apex rounded, sometimes
somewhat spatulate, the outer somewhat thickened at the top and
darker, mostly three-striate, inner scales sometimes slightly toothed
near the base; corolla light blue, clavate, 3-4 mm. long, teeth less
than I mm.; pappus white, spreading, about as long as the corolla;
achenia black, tapering at the base, three-, four- or five-angled, mostly
three, with the other angles obsolete, angles scabrous (359).

Collected in the coppet at Nicol's Town, March 17. Not un-
common. Same as Eggers 4424 and Brace 225.

Related to E. conyzoides Vahl, but differs in having the leaves and
branches less divaricate, leaves obtuse, involucral scales tapering at
the apex, heads fewer flowered, and achenia often 3-4-angled.

Plate 19. Eupatorium Bahamense. Portion of plant, x f ; a, head of flowers.

Eupatorium capillifolium (Lam.) Small. {E. foeniculaceum
Willd.) Hog Island, N.P., Feb. (252).

[em. Torrey Club, 12

Pl. 19

ENJATORIUM BAHAMENSE

HELIOTYPE CO., BOSTON.

FLORA OF NEW PROVIDENCE AND ANDROS 189

EuPATORiuM ODORATUM L. Nassau, Jan. (6). Same as Wright

295-

EuPATORiUM viLLOSUM Sw. Common and variable. Nassau,

Nicol's Town, March, Conch Sound, April (307, 340, 563). No.

307 from Nassau, and 563 from Conch Sound are almost entirely

glabrous forms.

WiLLUGHB^YA SCANDENS (L.) Kuntze. (Mikania Orinocensis
H.B.K.) Not uncommon in wet places Hog Island, Feb. ; Conch
Sound, May (245). Same as Wright 3600.

Aster exilis Ell. In damp places. Hog Island, N.P., Feb. ;
Conch Sound, April; Fresh Creek, June (246, 650).

Baccharis dioica Vahl. "Broom-bush." Common near the
shore. Nassau, Jan. (68).

Erigeron quercifolium Lam. In the pines. New Providence,
Feb.; Nicol's Town, March (324, 360).

Pluchea fcetida (L.) B.S.P. Damp ground in the pines.
Red Bays, April (498). Same as Eggers 4103.

Pluchea camphorata DC. Banana holes in the pines. N.P.,
Feb. (317).

Pluchea odorata (L.) Cass. "Cough-bush." New Provi-
dence, Feb, ; Nicol's Town, March (283).

Parthenium Hysterophorus L. Nassau, Jan, (i).

IvA CHEiRANTHiFOLiA Kth. Ncw Providcncc, Feb, (311). Same
as Eggers 4286.

IvA IMBRICATA Walt. Common on sandy shores. Deep Creek,
July (716).

Ambrosia HISPIDA Pursh. "Bay lavender," "sweet bay," Com-
mon on sandy beaches of both islands, Nassau, Jan. ; Mastic
Point, May (62).

IsocARPHA OPPOSiTiFOLiA R. Br. " Boston catnip." Deep
Creek, June (681). Determined at Kew by Dr. Britton.

BoRRiCHiA ARBORESCENS DC. Common on both islands.
Glabrate and canescent forms found growing together in several
localities. Nassau, Jan. (90). Canescent form the same as Eggers
1609, the glabrous the same as Wright 2899.

Amellus aspera (Jacq.) Kuntze. {Melanthera deltoidea Rich.)
Common on both islands. Nassau, Jan.; Red Bays, April (58,
503)-

190 FLORA OF NEW PROVIDENCE AND ANDROS

Salmea petrobioides Griseb. Nassau, Feb. Common along
shore (303).

BiDENS LEUCANTHA Willd. " Shepherd's needle." Nassau, Jan.

(36).

Tridax procumbens L. Nassau, Jan. (25).
Flaveria linearis Lag. Red Bays, April (462). Same as
Wright 2859.

PoROPHYLLUM RUDERALE Cass. Lisbon Creek, Andros, June

(675).

Pectis linifolia L. "Tea-blinkin." Nassau, Jan. (200).

Neurol^na lobata (L.) R. Br. Red Bays, April (486).

Erechthites hieracifolia (L.) Raf. Red Bays, April (504).

Emilia sonchifolia (L.) DC. Nassau, Jan. (3, 5).

Anastraphia Northropiana Greenman. "Candlewood."
Fresh Creek, June (743). Differs from the following species in
having the " leaves mostly entire, scales of the involucre more numerous
and strongly revolute." Named at Gray herbarium by Mr. J. M.
Greenman, 1897. Same as Combs no. 521 from Cuba, 1895.

Anastraphia pauciflosculosa Wright. "Candlewood." Lis-
bon Creek, June. Same as Eggers ^866. Determined at Kew
by Dr. Britton (639).

Chaptalia ALBICANS (DC). Conch Sound, March (400). Same
as Wright 2873.

SoNCHUS OLERACEUS L. Nassau, Jan. (48).

Relations of the Bahama Flora

Professor Hitchcock has treated this subject very fully in his
"Report on the Plants collected in the Bahamas, Jamaica, and Grand
Cayman." It is only taken up here because the exploration of Andros
has furnished additional data. In this connection it may be well to
describe briefly the character and position of the Bahama Islands in re-
lation to the Greater Antilles and North and South America. " The
Bahamas naturally divide themselves, first, into sunken banks like the
Navidad, Silver, and Mouchoir banks; next, islands occupying the
whole or nearly the whole summit of the banks from which they rise,
like Watlings, Rum Cay, Conception, Samana, Mariguana, the Plana
Cays, Inagua, Little Inagua, and the atoll of Hogsty; then banks
having the semblance of atolls, like the Crooked Island and Caicos

FLORA OF NEW PROVIDENCE AND ANDROS 191

banks, which are fringed by low islands forming a crescent with an
open lagoon or flat between its horns; next, Salt Cay Bank, which
from its structure holds a position intermediate between the group
of sunken banks like the Navidad and that resembling Caicos Bank,
and finally, such composite banks as the Little Bahama and Great
Bahama banks with the characteristics of a combination of banks
resembling all the others." ^

The Little Bahama Bank, lying in 26° to 27° north latitude,
is the most northerly. From it rise the islands of Great Bahama
and Abaco with a number of small cays. The Little Bahama Bank
is separated from the Great Bahama Bank lying south of it by the
Northeast and Northwest Providence channels, which are from
twenty to thirty miles wide and have a depth of from 500 to 2000
fathoms.

"The Great Bahama Bank is irregularly V-shaped and has an
extent of four hundred miles from northwest to southeast and is
about two hundred and fifty miles in its greatest width." ^

The water on the bank is usually only three or four fathoms
deep, but it is indented on the north by a tongue of the ocean which
extends nearly two thirds across it and has a depth of from
700 to 1200 fathoms. Along the western edge of this tongue
of ocean lies Andros, while New Providence is on the eastern
side, twenty-five miles or more distant. On the eastern border
of Great Bahama Bank lie the long narrow islands known
as Eleuthera, Cat Island, Exuma, and Long Island, the first two
being separated from the third by another indentation in the bank
from the south known as Exuma Sound. To the southeast of Cat
Island are the isolated islands of Rum Cay and San Salvador, or
Watlings Island, while east of the southern end of the Great Bahama
Bank is the much smaller bank on which are situated Crooked,
Acklin, and Fortune islands. Still farther southeast lie Mariguana,
the Caicos Bank and Islands and Turks Islands, while the entirely
isolated island of Inagua is off to the west. Inagua lies in a latitude
of about 21° and is the most southerly of the Bahamas. It is about
fifty miles from the east end of Cuba and about sixty miles north of
the western end of Haiti. From both islands it is separated by water
over 1500 fathoms in depth.

^ "A Reconnoissance of the Bahamas and of the Elevated Reefs of Cuba in the
Steam Yacht Wild Duck, January to April, 1893," Alexander Agassiz.

192 FLORA OF NEW PROVIDENCE AND ANDROS

The Little Bahama Bank is separated from Florida by a distance
of fifty miles and a depth of less than 450 fathoms of water. The
Great Bahama Bank extends west and northwest of Andros for a
distance of from fifty to seventy miles. At its western edge it is only
forty miles distant from Florida, and the channel is about 450 to 500
fathoms deep.

The Great Bahama Bank is separated from Cuba on the south
by the Old Bahama Channel, about 300 fathoms in depth, the narrow-
est part of which is only about twelve miles. Beyond this point the
water deepens rapidly to 500 and 1000 fathoms.

The following is quoted from Professor Hitchcock's admirable
account : " If, from any cause, the depth of the water of the ocean
should be lessened by 100 fathoms, there would be exposed the Little
Bahama and Great Bahama banks and several of the smaller banks
to the southeast. The Bahamas would be separated from the sur-
rounding islands and from Florida and the important channels would
still occupy the same places. If reduced by 300 fathoms, the Great
Bahama Bank would be united with Cuba. If the water were 500
fathoms shallower than at present, the Little and Great Bahama banks
would be united with Florida and some of the Windward Islands
would be connected. It is not, however, until a layer of water 1000
fathoms deep is removed that important changes would occur. Ja-
maica would be united with Honduras, Cuba with Florida and also
with South America through the Windward Islands. There would
be a narrow channel between Cuba and Yucatan, between Jamaica
and Haiti, and a wide and deep channel between Jamaica and Cuba.
Watlings, the Acklin Island group, and Inagua would still be isolated
and the distances between them and the neighboring land would not
be materially diminished. . . . The Greater Antilles are of ancient
formation and may have been connected with Mexico and Central
America at some remote period. But the Bahamas, the Windward
Islands, and the southern extremity of Florida are of recent origin."

Professor Hitchcock thinks, however, as does the writer, that
the ordinary methods of dissemination are sufficient to account
for the Bahaman flora and that the theory of an ancient land connec-
tion is not necessary.

The following table shows the distribution of the plants collected
by us on New Providence and Andros and bears out the conclusion

FLORA OF NEW PROVIDENCE AND ANDROS 193

arrived at by Hitchcock and others that the bulk of the Bahaman
plants have come from the south.

Although Andros is very nearly as close to Florida as it is to Cuba,
its plants are most closely allied with those of Cuba, comparatively
few species, apparently, having come from the north. Our explora-
tion of Andros, however, has shown that this northern contingent
is considerably larger than was supposed, a distinct colony of northern
plants having found a foothold on the west side of Andros. Many
of these have never been reported from either Cuba or any of the
other islands of the group.

It was to be expected that the majority of the Bahaman plants
would have a southern origin, on account of the greater similarity
of the climate and because the prevailing winds and currents are
from that direction. There are occasionally heavy "northers" during
the winter, when the wind blows hard from the northwest for a number
of days at a time. The velocity of the Gulf Stream as it flows through
the Florida Straits tends to prevent seeds being brought from the north
by water.

In compiling the table found below, giving the distribution of
the plants collected by us, the following volumes have been the
principal ones consulted: Grisebach's Flora of the British West
Indies, Grisebach's Catalogus Plantarum Cubensium, Plants col-
lected in the Bahamas by Hitchcock, Plants collected in the District
of Cienfuegos, Cuba, by Robert Combs, 1895; Jamaica, List of
Fawcett ; Flora of St. Croix and the Virgin Islands, Eggers ; Chap-
man's Flora of the Southern United States, and other works on the
North American flora.

194 FLORA OF NEW PROVIDENCE AND ANDROS

Tabulated Distribution

Name of Species

c/3

SCHIZ^ACE^

Ornithopteris adiantifolia (L.) Bernh.

I, 3> 4, 5-

Ornithopteris cicutaria (Kuntze) Un-

. derw.

POLYPODIACE^

Tectaria trifoliata (L.) Cav.

I, 3, 4, 5-

Dryopteris patens (Swz.) Kuntze.

I, 2, 3, 4, 5, widely dist.

Dryopteris asplenioides (Bak.) Kuntze

I, 3» 4, 5-

Davallia clavata Sw.

ij 3-

Asplenium dentatum L.

I, 3, 4-

Blechnum serrulatum Rich.

3, 4, 5-

Adiantum tenerum Sw.

I, 2, 3, 4, 5.

Pteridium caudatum (L.) Maxon.

I, 4, 5- .

Pteris longifolia L.

Widely distributed.

Vittaria lineata (L.) J. E. Smith.

I, 3y 4, 5> widely dist.

Cheilogramma lanceolata (L.) Blume.

I, 2, 3.

Acrostichum aureum L.

Widely distributed.

Phlebodium aureum (L.) R. Br.

I, 2, 4, 5.

Polypodium polypodioides (L.) Hitch.

Widely distributed.

Campyloneuron Phyllitidis (L.) Presl.

I, 2, 3, 4, 5.

Polypodium squamatum L.

Phymatodes Swartzii (Baker) Underw.

I. 3-

Goniopteris reptans (Swz.) Fee.

I, 4, 5-

Goniopteris reptans cordata.

PSILOTACE^

Psilotum nudum (L.) Griseb.

Widely distributed.

Cycadace^

Zamia sp.

1,6.

Conifers

Pinus Bahamensis Griseb.

Juniperus Barbadensis L.

I. 3-

Naiadace^

Ruppia maritima L.

Widely distributed.

Typhace^

Typha Domingensis Pers.

I, 2, 4, 5.

Gramine^

Paspalum fimbriatum H. B. K.

i» 3» 5-

Panicum divaricatum L.

Eleuthera, Cat, Crooked,

Panicum proliferum Lam.

3» 5-

Cenchrus tribuloides L.

Widely distributed.
Crooked, Inagua, i.

Sporobolus Virginicus Kth.

Widely distributed.
Crooked, Inagua.

Note. — i, Jamaica; 2, Virgin Islands; 3, Windward Islands; 4, Mexico and Central
America; 5, South America; 6, Haiti. When a plant is reported from all the preceding
localities (or all but Haiti) and also as being found in the Eastern Hemisphere, it is marked
" Widely distributed."

FLORA OF NEW PROVIDENCE AND ANDROS 195

Name of Species

Ph*

(/3

'z,

c/5

C/3

Gramine^. — Continued

Stenotaphrum Americanum Schrank.

Widely distributed.

Chloris Swartziana Doell.

I, 4, 5-

Chaetochloa glauca (L.) Scribn.

Widely distributed.
Inagua, Eleuthera.

Eragrostis ciliaris (L.) Link.

Widely distributed.

Eleuthera, Cat, Inagua, 5.

Uniola paniculata L.

Eleuthera, Cat, Crooked,
4, 5.

Cyperace^

Cyperus brunneus Sw.

Eleuthera, Cat, Crooked,
Inagua, i, 2, 3, 4.

Cyperus ferax Rich.

I, 3> 4, 5-

Cyperus ochraceus Vahl.

i> 3-

Eleocharis camptotriche Schweinitzii

Clarke.

Guadeloupe.

Eleocharis capitata (Willd.) R. Br.

Cat, I, 2, 5.

Eleocharis ochreata Nees.

Dichromena colorata (L.) A. S. Hitch.

Crooked, Fort., Inagua, 4, 5.

Fimbristylis spadicea (L.) Vahl.

Cat, Crooked, Fortune, In-
agua, I.

Fimbristylis monostachya (L.) Hassk.

Rynchospora cyperoides (Sw.) Mart.

Inagua, i, 3, 4, 5.

Rynchospora microcarpa Baldw.

Bahama, 5.

Cladium Jamaicense Crantz.

Crooked, 4, 5.

Scleria filiformis Sw.

Palm^

Thrinax Bahamense Cook.

Eleuthera, Cat, i, 2, 4, 6.

Paurotis Androsana Cook.

Cyclospathe Northropi Cook.

Bromeliace^

Tillandsia Balbisiana R. & S.

Tillandsia bulbosa Hook.

Cat, Inagua, i, 5.

Tillandsia fasciculata Sw.

Crooked, Inagua, i, 3, 4, 5.

Tillandsia flexuosa Sw.

Inagua, i, 5.

Tillandsia recurvata L.

Cat, I, 2, 3, 4, 5.

Tillandsia utriculata L.

Cat, Inagua, i, 2, 3, 5.

Commelinace^

Commelina nudiflora L.

Widely dist., i, 2, 3, 4, 5-

Rhoea discolor (L'Her) Hance.

I, 2, 3, 4.

LlLIACE^

Aletris bracteata Northr.

■ +

Smilace^e

Smilax auriculata Walt.

Smilax Havanensis Jacq.

2, 4, 5, 6.

Amaryllidace^

Agave rigida Mill.

Agave rigida Sisalana Engl.

Hymenocallis arenicola Northr,

Furcroea Cubensis Haw.

I, 2, 3, s.

Hypoxis juncea Smith.

196 FLORA OF NEW PROVIDENCE AND ANDROS

Name of Species

DlOSCORACE^

Rajania hastata L.

Orchidace^
Bletia verecunda R. Br.
Epidendrum fucatum Lindl.
Epidendrum nocturnum L.
Epidendrum Phoenicium Lindl.
Epidendrum odoratissimum Lindl.
Limodorum tuberosum L,
Stenorrhynchus orchioides (Rich.).
Gyrostachys tortilis (Rich.).
Oncidium sylvestre Lindl.
Oncidium variegatum Sw.
Govenia utriculata Lindl.
Polystachya luteola Hook.
Broughtonia lilacina Henfr.
Vanilla articulata Northr.

Casuarinace^
Casuarina equisetifolia Forst.

Myricace^
Myrica cerifera L.

MORACEiE
Ficus dimidiata Griseb.
Ficus pedunculata Willd.
Ficus pertusa L.

Ulmace^
Trema Lima (Lam.) A. S. Hitch.

Urticace^
Fleurya aestuans Gaud.
Adicea microphylla (Sw.) Kuntze.

LORANTHACEiE

Phoradendron racemosum Kr. & Urb.
Phoradendron Northropiae Urb.
Loranthus pauciflorus Sw.
Loranthus uniflorus Jacq.

Aristolochiace^
Aristolochia passiflorgefolia Rich.
Aristolochia pentandra L.

POLYGONACE^

Coccolobis diversifolia Jacq.
Coccolobis retusa Griseb.
Coccolobis tenuifolia L.
Coccolobis uvifera (L.) Jacq.

Coccolobis Wrightii Lindau.
Coccolobis obtusifolia Jacq.
Polygonum Portoricense Bertero.

CHENOPODIACEiE

Atrip! ex cristata H. B. K.

Cat, 2.

I, 4.

I, 3, 5.

i» 3» 5.

i» 3-

I, 2, 3, 6.

I, 5.

I, 2, 3, 5.

I, 6.

Crooked. Widely dist., nat.

Eleuthera, i.

I, 2.

Cat, I, 3, 4.

Eleuthera, i, 2, 3, 4, 5.

I, 2, 3, 4, 5-
I, 2, 3, 4, 5, 6.

I.
1,6.

I, 2, 3.

3-
I.

Cat, Watlings, Crooked,
Inagua, i, 2, 3, 4, 5-

I, 5.

FLORA OF NEW PROVIDENCE AND ANDROS 197

1, Name of Species

c/5

'z,

c/5

Chenopodiace^. — Continued

Salicornia ambigua Michx.

Crooked.

Salicornia Bigelovii Torr.

Dondia fruticosa (Forsk.).

Dondia linearis (Moq.) Millsp.

Amarantace^

Alternanthera muscoides Sw.

Inagua, 3.

Alternanthera paronychioides St. Hil.

Lithophila vermicularis (L.) Uline.

Eleuthera. Widely dist.

Iresine paniculata (L.) Kuntze.

Eleuth., Cat, Watlings
Widely distributed.

Batide^

Batis maritima L.

Inagua, i, 2, 3, 5.

Phytolaccace^

Rivina humilis glabra L.

I, 2, 3, 4, 5, Cat.

Phytolacca octandra L.

I, 3. 4, 5-

Nyctaginace^

Mirabilis Jalapa L.

Fortune, i, 3, 4, 5-

Boerhaavia erecta L.

I, 3» 4.

Boerhaavia paniculata Rich.

Eleuthera, i, 2, 3, 4.

Boerhaavia scandens L.

Eleuthera, i, 4, 5-

Pisonia aculeata L.

I, 3» 4, 5-

Pisonia obtusata Swz.

Crooked, i, 3, 5.

Pisonia rotundata Griseb.

AlZOACEiE

Sesuvium portulacastrum L.

Widely distributed. Eleu-
thera, Cat, Crooked,
Fortune.

PORTULACACE^

Portulaca oleracea L.

Eleuthera, Inagua. Widely
distributed.

Portulaca halimoides L.

I, 2, 3.

Anonace^

Anona palustris L.

I, 2, 3, 5.

Ranunculace^

Clematis dioica L.

I, 3, 4, 5.

Laurace^

Nectandra sanguinea Rottb.

I, 3, 4, 5. ^

Cassytha filiformis L.

Eleuthera, Cat, i, 2, 3, 4, 5-

Papaverace^e

Argemone Mexicana L.

Eleuth., Crooked, Fortune,
Inagua. Widely dist.

Crucifer^

Brassica arvensis (L.) B. S. P.

Int. from Europe, 2.

Lepidium Virginicum L.

Eleuthera, Cat, Inagua,
I, 2, 3, 4.

Cakile aequalis L'Her.

I, 2, 3.

Capparidace^

Pedicellaria pentaphylla (L.) Schrank.

+ +

Eleuthera. Widely dist.

198 FLORA OF NEW PROVIDENCE AND AND R OS

Name of Species

1:3

"it

lyj

Crassulace^

Bryophyllum pinnatum (Lam.) S.

Widely distributed.

Kurz.

Rosacea

Chrysobalanus Icaco L.

Widely distributed.

MlMOSACE^

Acacia choriophylla Benth.

Acacia Farnesiana (L.) Willd.

Widely distributed. For-
tune, Inagua, i, 3, 5.

Acuan depressa (Kth.) Kuntze.

Inagua, i, 2, 3, 4, 5-

Acuan virgata (Willd.) Kuntze.

Eleuthera. Widely dist.

Mimosa pudica L.

Widely distributed.

Leucaena glauca (L.) Benth.

Widely distributed. Eleu-
thera, Cat, Fortune.

Lysiloma paucifoliola (DC.) Hitch.

Eleuthera, Fortune.

Lysiloma Bahamensis Benth.

Cat, Fortune, Inagua.

Calliandra formosa Benth.

Pithecolobium Hystrix Benth.

Pithecolobium Unguis-cati (L.) Bth.

Eleuthera, Cat, Fort., i, 2.

Pithecolobium Bahamense Northr.

C^SALPINACE^

Cassia Bahamensis Mill.

Eleuthera, Cat, Watlings,

Cassia biflora L.

Eleuthera, Fort., Inagua,

I, 2, 3, 4, 5, 6.

Cassia ligustrina L.

I, 5-

Cassia mimosoides L.

Eastern Hemisphere.

Cassia occidentalis L,

Widely distributed. Eleu-
thera, Fortune, Inagua.

Cassia polyadena DC.

Cat, Crooked, i, 3.

Cassia villosa Mill.

Cassia Caribsea Northr.

Tamarindus Indica L.

Inagua, i.

Haematoxylon Campechianum L.

I, 2.

Csesalpinia ovalifolia Urb.

Caesalpinia crista L.

Eleuthera, Inagua,

Caesalpinia Rugeliana L.

Inagua, 3.

Peltophorum adnatum Griseb.

Papilionace^

Sophora tomentosa L.

Crooked, Fortune, Inagua.
Widely distributed.

Crotalaria pumila Ort.

Eleuthera, Cat, i, 2, 3, 4, 5-

Crotalaria retusa L.

Widely distributed.

Crotalaria verrucosa L.

I, 2, 3 (from Old World).

Indigofera Anil L,

Eleuthera, Inagua. Widely
distributed.

Cracca Schottii Vail.

Asia.

Stylosanthes hamata (L.) Taub.

Eleuthera, Fortune, Inagua.
Widely distributed.

Meibomia incana (Sw.) Kuntze.

Eleuthera, Cat. Widely dist.

Abrus precatorius L.

I, 2, 3, 5. Widely dist.

Bradburya Virginiana (L.) Kuntze.

Eleuthera, Cat, Crooked,
Inagua. Widely dist.

Bradburya angustifolia (DC.) Griseb.

Cat. Widely distributed.

FLORA OF NEW PROVIDENCE AND ANDROS 199

Name of Species

Ph'

c/i

Papilionaceae. — Continued

Galactia rudolphioides (Griseb.) Wr.

Eleuthera, Cat, Crooked.

Galactia Cubensis H. B. K.

Watlings, Inagua. Widely
distributed.

Canavalia obtusifolia (Lam.) DC.

Eleuthera, Inagua. Widely
distributed.

Rhynchosia minima (L.) DC.

Inagua. Widely dis't.

Phaseolus semierectus L.

Widely distributed.

Vigna repens (L.) Kuntze.

I, 2, 3, 4, 5-

Cajanus Cajan (L.) Millsp.

Eleuthera. Widely dist.

Dalbergia Brownei (Pers.) Kuntze.

Widely distributed.

Ichthyomethia Piscipula (L.) Hitch.

Cat, I, 2, 3, 4, 5.

OXALIDACE^

Oxalis corniculata repens Zucc.

Widely distributed.

Linages

Linum Bahamense Northr.

Erythroxylon brevipes DC.

Fortune, i.

Erythroxylon obovatum Macf.

i» 3-

Erythroxylon reticulatum Northr.

Malphigiace^

Byrsonima lucida Rich.

2, 3-

Malphigia setosa Spr.

Cat, 6.

Stigmaphyllon Sagraenum Juss.

Triopteris rigida Sw.

Rutace^

Xanthoxylon cribrosum Spr.

Xanthoxylon emarginatum Sw.

Cat, Inagua, i.

Fagara Fagara (L.) Small.

Eleuthera, Cat, i, 3, 4.

Simarubace^

Suriana maritima L.

Eleuthera, Cat, Crooked, 2.

Picrodendron baccatum Bahamense.

BURSERACE^

Bursera Simaruba (L.) Sarg.

Eleuthera, Cat, Crooked,
I, 2, 3, 4, 5-

Swietenia Mahogani L.

Crooked, Inagua, i, 2, 4, 5.

POLYGALACE^

Polygala Boykinii Nutt.

Polygala brizoides St. Hil.

I, 4, 5.

Polygala spathulata Griseb.

EUPHORBIACE^

Phyllanthus Bahamensis Urb.

Phyllanthus Epiphyllanthus L.

Eleuthera, Cat, Watlings,
Fortune, Inagua, i, 2.

Phyllanthus Niruri L.

Fortune, Inagua. Widely

Croton linearis Jacq.

Eleuthera, Cat, i. [dist.

Savia erythroxyloides Griseb.

Acalypha alopecuroides Jacq.

Eleuthera, 4, 5-

Lasiocroton macrophyllus Griseb.

Excoecaria lucida Sw.

I, 3-

Exccecaria Sagraei J. Mull.

200 FLORA OF NEW PROVIDENCE AND ANDROS

Name of Species

C/5

EuPHORBiACE^. — Continued

Hippomane Mancinella L.

Inagua, i, 2, 3, 4, 5-

Bonania emarginata DC.

Euphorbia Blodgettii Engelm.

Eleuthera, Cat, Crooked
Fortune, Inagua, i.

Euphorbia buxifolia Lam.

Eleuthera, Cat, Watlings,
Crooked, Inagua, i, 2, 3,

Euphorbia cassythoides Boiss.

A, 5-

Euphorbia heterophylla L.

Eleuthera, Cat, Crooked,
Inagua, i, 2, 4, 5.

Euphorbia heterophylla graminifolia

Engelm.

Eleuthera, Cat, Fortune,
Inagua, i, 2, 4, 5.

Euphorbia nutans Lag.

Widely distributed. Eleu-
thera, Inagua.

Euphorbia serpens Kth.

3, 5-

Buxus Bahamensis Baker.

Anacardiace^

Metopium Metopium (L.) Small.

Cat, Crooked, Inagua, i.

Celastrace^

Elaeodendron xylocarpum DC.

Bermudas.

Maytenus buxifolius (Rich.) Griseb.

Cat, Crooked.

Crossopetalum pallens (Smith)

2, 3-

Crossopetalum coriaceum Northr.

Schaefferia frutescens Jacq.

Eleuthera, i, 2, 3, 4, 5-

iLICACEiE

Ilex sideroxyloides Griseb.

Sapindace^

Serjania diversifolia Radlk.

Serjania subdentata Juss & Poir.

Eleuthera, Cat, i, 2, 3.

Cardiospermum Halicacabum L.

Widely distributed.

Thouinia discolor Griseb.

Eleuthera, Cat, Fortune,
Inagua.

Exothea paniculata (Juss.) Radlk.

Hypelate trifoliata Sw.

Alvaradoa amorphioides Liebm.

Rhamnace^

Krugiodendron ferreum (Vahl.) Urb.

I, 2, 3.

Colubrina ferruginosa Brongn.

I, 2, 3.

Reynosia latifolia Griseb.

Eleuthera, Cat, 2

Reynosia Northropiana Urb.

Gouania Domingensis L.

+ .

I, 2, 3, 5.

Vitace^.

Vitis rotundifolia Michx.

Vitis alata Jacq.

Vitis trifoliata (L.) Bak.

I, 2, 3, 5.

Cissus sicyoides L.

I, 2, 3, 4, 5-

Parthenocissus quinquefolia (L.)

Planch.

TlLIACE^

Triumfetta semitriloba L.

Widely distributed.

FLORA OF NEW PROVIDENCE AND ANDROS 201

Name of Species

S
I

C/2

■*f.

c/5

TiLiACE^. — Continued

Corchorus hirsutus L.
Corchorus siliquosus L.

+
+

Eleuthera, Cat, Crooked,

Watlings, Inagua, i, 2, 3.

Eleuthera, i, 2, 3, 4, 5, 6.

Malvace^

Sida carpinifolia L.

Eleuthera, Inagua. Widely

distributed.
2, 3, 4, 5, 6.
Eleuthera, 3, 4, 5.
I, 2, 3, 5.

Widely distributed.

Sida supina L'Her,
Abutilon crispum Don.
Pavonia spicata Cav.
Hibiscus cryptocarpus Rich.
Hibiscus tiliaceus L.

+
+
+

+
+

+
+
+

Sterculiace^

Helicteres semitriloba Bert.
Helicteres spiralis Northr.
Melochia nodiflora Sw.
Melochia tomentosa L.

+
+

Inagua, 6.

I, 2, 3, 4, 5-

Eleuthera, Cat, Inagua, i,

Waltheria Americana L.
Hypericace^

2, 3, 4, 5-
Eleuthera, Cat, Fortune.
Widely distributed.

Ascyrum hypericoides L.

I, 4, 5-

Bixace^

Xylosma ilicifolium Northr.

Canellace^

Canella Winterana (L.) Gaertn.

I, 2, 3.

Turnerace^

Turnera ulmifolia L.

Eleuthera, Cat, Fortune,

Passiflorace^

i> 2, 3, 4, 5-

Passiflora angustifolia Sw.
Passiflora cupraea L.
Passiflora minima L,
Passiflora multiflora L.
Passiflora pectinata Griseb.

+
+
+
+
+

+
+
+
+

+
+

Eleuthera, Cat.

Eleuthera, Inagua, i, 6.

Turk.

Cactace^

Cereus Swartzii Griseb.
Opuntia spinosissima Mill.
Opuntia Tuna (L.) Mill.

+
+
+

Cat, Crooked, Inagua, i.
Fortune, Inagua, i, 2.
Crooked, Inagua, i, 2,3,4,5.

LyTHRACEtE

Parsonsia Parsonsia (L.) Britton.

Rhizophorace^

Rhizophora Mangle L.

Cat, Crooked.Widely dist.

Myrtace^

Calyptranthes pallens Griseb.
Myrtus punctata Griseb.
Eugenia axillaris (Sw.) Willd.
Eugenia longipes Berg.
Eugenia monticola DC.

+
+
+
+
+

+
+
+

1, 2, 3.

2, 3, 6.
I, 2.

Eleuthera, i, 2, 3.

202 FLORA OF NEW PROVIDENCE AND ANDROS

Name of Species

COMBRETACEiE

Conocarpus erecta L.

Laguncularia raceniosa (L.) Gaertn,
Terminalia spinosa Northr.

+
+

Eleuthera, Cat, Watlings,
Crooked, Inagua.
Widely distributed.

Fortune, Crooked. Widely
distributed.

Melastomace^

Tetrazygia bicolor (Mill) Cogn.

Onograce^

Jussiaea sufifruticosa L.

Widely distributed.

Samydace^

Casearia laetioides (Rich.).
Banara reticulata Griseb.

+
+

I, 4.

Umbellifer^

Hydrocotyle pygmsea Wright.
Centella Asiatica (L.) Urb.

Widely distributed.

Myrsinace^

Rapania Guyanensis (Aubl.).
Icacorea paniculata (Nutt.) Sudw.
Jacquinia Keyensis Mez.

Primulace^

+
+

+
+
+

I, 3» 4, 5-
4-

Eleuthera, Cat, Crooked,
I, 2, 3, 4, 5, 6.

Samolus ebracteatus Kth.

Plumb AGIN ace^

Plumbago scandens L.

Inagua, i, 2, 3, 4, 5-

SAPOTACEiE

Chrysophyllum oliviforme Lam.
Lucuma pauciflora A. DC.
Sideroxylon mastichodendron Jacq.
Bumelia microphylla Griseb.
Bumelia Cubensis Griseb.
Dipholis salicifolia A. DC.
Mimusops dissecta R. Br.
Mimusops Floridana Engelm.

+
+

+
4-
+
+
+
+
+
+

+
+
+

+
+

+
+
+

I, 2, 6.
I, 2, 3, 6.

I, 2, 3.
3-

Ebenace^

Maba Caribaea (A. DC.) Hiern.

Oleace^

Adelia porulosa (Poir.) Engler.

I, 2, 4-

LOGANIACE^

Spigelia Anthelmia L.
Cynoctonum mitreola (L.) Britt.
Cynoctonum sessilifolia (T. & G.)
Britt.

Gentianace^

+
+

Eleuthera, Fort., i, 3, 4, 5-
I, 3» 4, 5, 6.

Voyria Mexicana Griseb.
Eustoma exaltatum Griseb.
Sabbatia campanulata (L.) Torr.

+
+

+
+
+

+
+

I, 4, 5, 6.

FLORA OF NEW PROVIDENCE AND ANDROS 203

Name of Species

Apocynace^
Vinca rosea L.
Plumiera obtusa L.
Echites Andrewsii Chapm,

Echites biflora Jacq.
Echites Sagrsei A. DC.
Echites umbellata Jacq.

ASCLEPIADACE^

Asclepias Curassavica L.
Metastelma Bahamense Griseb.
Metastelma barbatum Northr.
Metastelma palustre Schltr.

CONVOLVULACE^

Ipomoea cathartica Poir.
Ipomcea coccinea L.
Ipomoea commutata R. & S.
Ipomoea fastigiata Sweet.
Ipomoea heptaphylla Griseb.
Ipomoea Jamaicensis Don.
Ipomoea grandiflora Lam.
Ipomoea Pes-caprae (L.) Sweet.

Ipomoea repanda Jacq.
Ipomoea sidifoHa Chois.
Ipomoea sinuata Ort.
Ipomoea triloba L.

Tacquemontia Jamaicensis (Jacq.)
Hall. ^^ ^^

Jacquemontia verticillata (L.) Urb.
Evolvulus arbuscula Poir.

Dichondra repens Forst.

CuSCUTACEiE

Cuscuta Americana L.

Hydrophyllace^
Nama Jamaicensis L.

Boraginace^
Cordia angustifolia R. & S.
Cordia lima R. & S.
Cordia Sebestena L.
Bourreria Havanensis (Willd.) Miers.

Tournefortia gnaphalodes (Jacq.)

R. Br.
Tournefortia volubilis L.
Heliotropium Curassavicum L.

Heliotropium parviflorum L.

Heliotropium nanum Northr.

Widely distributed.

Cat, I, 2, Fortune, Inagua.

Eleuthera, Cat, Watlings,

I, 2, 3, 4.
3, 5-

Eleuthera, Cat, Fortune,
Crooked, Inagua, i, 4, 6.

Widely distributed.
Fortune, Inagua, 2.

I, 3, 4, 5-

Widely distributed.

I, 5-

h 2, 3, 4, 5-

Eleuthera, i, 4, 5.

Fortune, Inagua. Widely

distributed.
Eleuthera, Cat, Fortune, 3.
I, 4, 6.

Eleuthera. Widely dist.
Eleuthera, Fortune, Inagua,

Crooked. Widely dist.
Eleuthera, Cat, Crooked,

Inagua, i, 6.

I, 3» 5-

Eleuthera, Fortune, Inagua,

I, 6.
Widely distributed.

Eleuthera, i, 2, 3, 4, 5.
I, 4, 6.

Inagua, i, 3, 5.

Eleuthera, Cat, Crooked,
Fortune, i.

Eleuthera, Watlings, In-
agua, I, 2, 3.

Eleuthera, Inagua, i, 2, 3, 5.

Cat, Crooked, Inauga.
Widely distributed.

Eleuthera, Cat, Inagua, i,
2, 3, 4, 5-

204 FLORA OF NEW PROVIDENCE AND ANDROS

Name of Species

d
E

c/5

C/3

c/j

Verbenace^

Lantana Camara L.

Watlings, i, 2, 3, 4, 5, 6.

Lantana crocea Jacq.

Eleuthera, Cat, i, 3, 5, 6.

Lantana involucrata L.

Eleuth., Watlings, Crooked,
Cat, Inag., i, 2, 3, 4, 5-

Lippia canescens Kth.

Lippia nodiflora (L.) Michx.

Eleuthera, Cat, Inagua.
Widely distributed.

Abena Jamaicensis (L.) A. S. Hitch.

Eleuthera, Cat, Crooked,
Inauga. Widely dist.

Citharexylum Berterii Spreng.

Citharexylum lucidum Cham. & Schl.

I, 3» 4, 5-

Duranta repens L.

Eleuthera, i, 2, 3, 4, 5.

Petitia Domingensis Jacq.

Grand Cayman, i, 6.

Vitex ilicifolia Rich.

Avicennia nitida Jacq.

Fortune. Widely distributed.

Labiate

Micromeria Brownei Benth.

I, S-

Salvia occidentalis Sw.

i> 3> 4, 5-

Salvia serotina L.

Eleuthera, Cat, Watlings,
Crooked, Inag., i, 2, 3, 4.

Leonurus Sibiricus L.

Widely distributed.

Leonotis nepetaefolia (L.) R. Br.

Eleuthera, Inagua. Widely
distributed.

Teucrium Cubense L.

Cat, Inagua, 4, 5-

Mesosphaeriim pectinatum (Poit.)

Kuntze.

Inagua. Widely dist.

SOLANACE^

Solanum aculeatissimum Jacq.

I, 2, 4, 5-

Solanum Bahamense L.

Eleuthera, Cat, Watlings
Crooked, i, 2, 3, 4.

Solanum nigrum L.

Inagua, i, 2, 4.

Solanum nigrum nodiflorum Gray.

Inagua, i, 3.

Solanum verbascifolium L.

Eleuthera. Widely dist.

Physalis angulata L.

I, 2, 3, 5.

Physalis Barbadensis Jacq.

Widely distributed.

Capsicum baccatum L.

Cestrum pallidum Lam.

Cat, I, 6.

Datura Metel L.

Inagua. Widely dist.

SCROPHULARIACE^

Stemodia maritima L.

Eleuthera, i, 5.

Capraria biflora L.

Cat, Watlings, Crooked,
Inagua. Widely dist.

Buchnera elongata Sw.

I, 3» 4, 5-

Gerardia maritima Raf.

Gerardia purpurea L.

Lentibulariace^

Pinguicula pumila Michx.

BlGNONIACE^

Tecoma lepidophylla Griseb.

Tecoma Bahamensis Northr.

Tecoma stans (L.) Juss.

Inagua, i, 2, 3, 4.

Jacaranda Bahamensis R. Br.

FLORA OF NEW PROVIDENCE AND ANDROS 205

Name of Species

c/5

c/j

BiGNONiACE^. — Continued

Jacaranda coerulea Griseb.

Cat.

ACANTHACE^

Thunbergia fragrans Roxb.

I, 3 (East Indies).

Blechum Brownei Juss.

Widely distributed.

Anthacanthus acicularis (Sw.) Nees.

Dicliptera assurgens Juss.

I, 3, 4, 5-

RUBIACEiE

Exostemma Caribaeum (Jacq.) R. & S.

Crooked, i, 2, 3, 4, 5.

Rhacicallis Americana (Jacq.) Hitch.

Crooked, i.

Catesbaea spinosa L.

Catesbaea fasciculata Northr.

Randia aculeata L.

Eleuthera, Cat, Crooked,
Inagua, i, 2, 3, 6.

Genipa clusiaefolia (Jacq.) Griseb.

Cat, Watlings, Crooked,
Fortune.

Guettarda elliptica Sw.

Eleuthera, Fortune, Inagua,

Guettarda scabra Lam.

I, 4.
Inagua, i, 2, 3, 4, 5.

Antirrhoea lucida Gaertn.

I, 2, 3, 6.

Antirrhoea myrtifolia Griseb.

Erithalis fruticosa L.

Eleuthera, Cat, Watlings,
Fort., Crooked, Inagua,
i> 2, 3, 4.

Erithalis rotundata Griseb.

Chiococca parvifolia Wulls.

Crooked, i, 3, 4, 5.

Chiococca alba (L.) A. S. Hitch.

Eleuthera, Cat, i, 2, 3, 4, 5-

Chiococca sp.

Phialanthus myrtilloides Griseb.

Crooked.

Strumpfia maritima Jacq.

Eleuth., Watlings, Crooked,

Morinda Royoc L.

3-
I, 2, 6.

Myrstiphyllum pubescens (Sw.) A. S.

Hitch.

I, 4.

Myrstiphyllum undatum (Jacq.) Hitch

Eleuthera, Cat, i.

Myrstiphyllum ligustrifolium Northr.

Ernodea littoralis Sw.

Cat, Watlings, Crooked,
Fortune, Inagua, i, 3.

Hamelia patens Jacq.

I, 2, 3, 4, 5-

Spermacoce tenuior Lam.

Cat, Crooked, Inagua, For-
tune, Eleuth., I, 2, 3, 4, 5.

Galium hispidulum Mx,

CUCURBITACE^

Anguria Keithii Northr.

Trianosperma racemosa (Griseb.) T.

&G.

I, 5, 6.

GOODENIACE^

Scaevola Plumieri (L.) Vahl.

Eleuthera, Crooked.
Widely distributed.

Composite

Vernonia Bahamensis Griseb.

Cat, Inagua.

Ageratum conyzoides L.

Fortune, Inagua. Widely
distributed.

206 FLORA OF NEW PROVIDENCE AND ANDROS

Name of Species

c/j

(72

Composite. — Continued

Eupatorium ageratifolium DC.

4,6.

Eupatorium Bahamense Northr.

Eupatorium capillifolium (Lam.)

Small.

Eupatorium odoratum L.

I, 3, 4, 5, 6.

Eupatorium villosum Sw.

Cat, I.

Willughbaeya scandens (L.) Kuntze.

I, 3> 4, 5-

Aster exilis Ell.

Baccharis dioica Vahl.

Eleuthera, Inagua, i, 2, 3.

Erigeron quercifolium Lam.

Pluchea foetida (L.) B. S. P.

Pluchea camphorata DC.

I, 3, 4, 5-

Pluchea odorata (L.) Cass.

Inagua, i, 2, 3, 4, 5-

Parthenium Hysterophorus L.

Eleuthera, Cat, Inagua, i,
2, 3, 4, 5-

Iva cheiranthifolia Kth.

Iva imbricata Walt.

Ambrosia hispida Pursh.

Eleuthera, Crooked, For-
tune, I, 4.

Isocarpha oppositifolia R. Br.

Borrichia arborescens DC.

Cat, Crooked, i, 2, 3, 5.

Amellus aspera (Jacq.) Kuntze.

Eleuthera, Fortune, i, 2,

3, 4, 5-
Grand Cayman.

Salmea petrobioides Griseb.

Bidens leucantha Willd.

Eleuthera. Widely dist.

Flaveria linearis Lag.

Porophyllum ruderale Cass.

I, 3, 4, 5-

Pectis linifolia L.

I, 2, 3.

Neurolaena lobata (L.) R. Br.

I, 3> 4, 5-

Erechthites hieracifolia (L.) Raf.

Widely distributed.

Emilia sonchifolia (L.) DC.

Widely distributed-

Anastraphia Northropiana Greenm.

Anastraphia pauciflosculosa Wright.

Fortune, Crooked, Inagua,
6.

I, 4, 5-

Chaptalia albicans (DC.) H.

Sonchus oleraceus L.

Eleuthera. Widely dist.

Tridax procumbens L.

Summary of Above Table

Total found on New Providence and Andres (exclusive of cultivated and

escaped plants), 453

Of these there have been reported

From other islands of the group, 176

From Cuba, 335

From South Florida, 250

From southern United States, 108

From Jamaica, 286

From Virgin Islands, 19°

From Windward Islands, 223

From Mexico and Central America, 19 7

From South America, i99

FLORA OF NEW PROVIDENCE AND ANDROS 207

Of these 453 species, 76 are widely distributed, being common
in warm countries on both continents.

Distribution of the 128 Species found in New Providence and An-
dros, but not reported from cuba

Peculiar to the Bahamas, 34

Found in south Florida, 47

Found in southern United States, 28

Found in Jamaica, 20

Found in Virgin Islands, 10

Found in Windward Islands, 14

Found in Mexico and Central America, 8

Found in South America, 10

Found in Haiti, 10

The 34 species that are apparently endemic are as follows : Pinus
Bahamensis, Acacia choriophylla, Buxus Bahamensis, Passiflora
pectinata, Jacaranda Bahamensis, Jacaranda ccerulea, Vernonia
Bahamensis, Thouinia discolor, Ccesalpinia ovalifolia, Phyllanthus
Bahamensis, Casearia Bahamensis, and Reynosia Northropiana;
and the following described and figured as new in this report : Thrinax
Bahamensis, Paurotis Androsana, Cyclospathe Northropi (the last
two new genera), Hymenocallis arenicola, Vanilla articulata, Pithe-
colobium Bahamense, Cassia Carihcea, LinumBahamense, Erythroxylon
reticulatum, Phorodendron Northropice, Helicteres spiralis, Xylosma
ilicifolia, Terminalia spina sa, Tecoma Bahamensis, Cateshcea fasci-
culata, Myrstiphyllum ligustifolium, Crassopetalum coriaceum, Meta-
stelma harhatum, Heliotr opium nanum, Aletris hracleata, Anguria
Keithii, Eupatorium Bahamense. Of the 34 species, 21 were found
only on Andros.

Of the other species formerly considered endemic (Report of
Committee appointed for Purpose of Exploring the Flora of the Ba-
hamas, W. T. Thiselton-Dyer, Sec, 1888) Phialanthus myrtilloides
and Antirrhcea myrtifolia have since been reported from Cuba ; Salmea
petrohioides has been reported by Hitchcock as occurring in Grand
Cayman and Bletia purpurea has proved to be not distinct from B.
verecunda.

Distribution of the 250 Species found in New Providence and

Andros, and also in South Florida

Found in the United States, north of tropical Florida, 108

Found in Cuba, 202

208 FLORA OF NEW PROVIDENCE AND ANDROS

Found in Jamaica, 198

Found in Virgin Islands, 149

Found in Windward Islands, 161

Found in Mexico and Central America, 153

Found in South America, 153

Widely distributed, 68

Species not hitherto reported outside of the United States, 18
Of these 18 species, 4 are limited to tropical Florida, the other 14
extend north of southern Florida.

Of the plants apparently restricted to the United States and the
Bahamas, Dondia linearis was found only on New Providence;
Smilax auriculata, Vitis rotundifolia, Sabbatia campanulata, Eri-
geron quercifolium, and Jacquinia Keyensis were collected on both
islands while the following were found only on the island of Andros :
Iva imbricata, Eugenia longipes, Mimusops Floridana, Myrica ceri-
fera, Plucheafoetida, Salicornia Bigelovii, Poly gala Boykinii, Samolus
ebracteatus, Cynoctomum sessilifoliaj Gerardia maritima, Gerardia pur-
purea, Pinguicula pumila. The last eight were confined to the
western side of Andros, growing on the savannas, the border of the
swash, or, in the case of the Pluchea, in damp spots in the pines.
Myrica cerifera is said by Gardiner and Dolley to have been intro-
duced from the United States.

The three reported by Hitchcock as being confined to the United
States and the Bahamas are Xanthium strumarium, Vitis rotundifoliay
and Distichlis spicata (the last found only in Inagua). The Vitis j
he observes, may have been carried by birds, and the Smilax, Eugenia,
and Mimusops are probably due to the same agency. Many of our
seed-eating birds either spend their winters in the Bahamas or stop
there on their migrations. Catbirds and mocking birds, for instance,
were abundant on Andros during the winter and early spring.

The plants mentioned above as being found on the savannas and
bordering the swash on the west side of Andros offer no inducements,
however, to seed-eating birds, the fruits being dry and inconspicuous
and the seeds in many cases minute. As has been noted before, the
west side of Andros is a paradise for water birds, and they are found
there in large numbers. Many of these birds, such as the great blue
heron {Ardea herodias), the little blue heron (Ardea ccRrulea), and the
killdeer (^Egialitis vocifera), are regular winter visitors from the United
States. May it not be that some of these waders have at some

FLORA OF NEW PROVIDENCE AND ANDROS 209

time transported seeds of the plants in question, in mud that may
have adhered to their beaks or feet ? Their presence certainly seems
very difficult to account for otherwise. Darwin says ("Origin of
Species," chapter thirteen): "I have before mentioned that earth
occasionally adheres in some quantity to the feet and beaks of birds.
Wading birds which frequent the muddy edges of ponds if suddenly
flushed would be the most likely to have muddy feet. Birds of this
order wander more than those of any other, and they are occasionally
found on the most remote and barren islands of the open ocean;
they would not be likely to alight on the surface of the sea, so that any
dirt on their feet would not be washed off, and when gaining the land,
they would be sure to fly to their natural fresh-water haunts. I do
not believe that botanists are aware how charged the mud of ponds
Js with seeds. I have tried several little experiments, but will here
give only the most striking case : I took in February three tablespoon-
fuls of mud from three different points, beneath water on the edge of
a little pond ; this mud when dried weighed only 6J ounces ; I kept
it covered up in my study for six months, pulling up and counting
each plant as it grew ; the plants were of many kinds and were alto-
gether 537 in number, and yet the viscid mud was all contained in a
breakfast cup ! Considering these facts, I think it would be an inex-
plicable circumstance if water birds did not transport the seeds of
fresh-water plants to unstocked ponds and streams situated at very
distant points." The plants mentioned above are not water plants,
it is true, but they are common in moist soil in the vicinity of ponds.
The seeds of Pluchea may owe their transportation to the wind.
But whatever the means of dissemination, the fact seems established
that although the bulk of the Bahaman flora has probably come
from the south, there is a contingent, in the northern islands of the
group at least, that owes its origin to the north. It is worthy of note
in this connection that in a number of cases, when our plants were
compared with large series of both Florida and Cuban specimens,
they were found to resemble most closely the Florida specimens;
hence when species occur in both Cuba and Florida, it may well be
that the Bahaman plants owe their origin to the latter.

Distribution within the Bahaman Group
Species found on Andros, 359

Species found on New Providence, 262

210 FLORA OF NEW PROVIDENCE AND ANDROS

Species common to both islands, 153

Species reported from Eleuthera, 98

Cat Island, 79

Crooked and Fortune islands, 79

Inagua, 88

Watlings, 18

(The above data for plants reported from the islands other than
New Providence and Andros were taken from Hitchcock's Report.)

In addition to the plants collected by us in New Providence
and Andros, Professor Hitchcock lists in his report 148 more, col-
lected from the various islands of the group. Of these 36 are grasses
and 30 are widely distributed or introduced species, while Epidendrum
altissimum, Mimosa Bahamensis, Croton Eleuteria, Croton Hjalmar-
sonii, Pavonia Bahamensis, and Eragrostis Bahamensis are endemic,
making a total of forty endemic species in the Bahama Islands. The
last two had been previously undescribed.

Notes on the Local Distribution

The Bahama pine {Pinus Bahamensis), so abundant on New
Providence and Andros, is confined entirely to the northern islands
of the group, being found in addition only on Abaco, Bahama, and
the Berry islands, the first two being on the Little Bahama Bank, and
the latter a series of small cays on the Great Bahama Bank north of
Andros.

Although, as we have seen, New Providence and Andros have
many plants in common, some interesting points of difference were
noted. The numerous introduced plants so common about Nassau,
as Argemone Mexicana, Asclepias Curassavica, Abrus precatorius,
Bidens leucantha, Ageratum conyzoides, etc., were, of course, entirely
wanting on Andros, but, on the other hand, Aristolochiaceae, Cacta-
ceae, Polygalaceae, and Loranthaceae were not seen on New Provi-
dence while represented by two or three species each on Andros.
Passiflora pectinata, while very common in the pines on Andros (also
reported from Turk's Island by Grisebach), was not found on New
Providence, as was also the case with Hymenocallis arenicola, the
latter growing luxuriantly on many of the sandy beaches of the
former island.

Among the interesting plants found on Andros that have not

FLORA OF NEW PROVIDENCE AND ANDROS 211

been reported from New Providence are the following: Agave
rigida, Casearia Icetioides, Pithecolohium Hystrix, PeUophorum adna-
tum, Ichthyomethia Piscipula, Euphorbia cassythoideSj Lucuma
pauciflora, Voyria Mexicana, Ipomcea re panda, Petitia Domingensis,
Vitex ilicifolia, Mimusops Floridana, and Morinda Royoc. With
very few exceptions the plants found on the savannas of Andros
were wanting on New Providence, and as they formed the most in-
teresting feature of the Andros flora, it may be worth while to give
the full list of plants found there. They were Sahhatia campanulata,
Eustoma exaltatum, Bletia verecunda, Euphorbia buxifolia, Dichromena
colorata, Cladium Jamaicense, Merosphcerum pectinatum, Gyrostachys
tortilis, Flaveria linearis, Hypoxis juncea, Cynoctomum sessilifolia
and C. Mitreola, Pinguicula pumila, Buchnera elongata, Gerardia pur-
purea and G. maritima, Poly gala Boykinii and P. brizoides, Samolus
ebracteatus, Limodorum tuberosum, Aletris bracteata, Chiococca alba
parviflora, Buxus Bahamensis, Heliotropium nanum, Crassopetalum
coriaceum, Rhamnidium punctatum, and Terminalia spinosa. Of
these only the first six were collected on New Providence, and in
addition to Rhamnidium punctatum, were the only ones found on
Andros elsewhere than on the savannas.

On Andros itself, no Cactaceae were noted in the northern part
of the island, while Cereus Swartzii and two species oi Opuntia were
quite common in the southern part, as they are said to be on the more
southerly islands of the group. We also noticed that, whereas Vitis
alata was very common on the northern half of the island, below that
it was replaced by the allied species Vitis trifoliata.

On both islands mosses, lichens, and fungi were noticeably
infrequent, and of the few species of each that were found none were
common.

CULTIVATION OF SISAL IN THE BAHAMAS^
By John I. Northrop

''Are you interested in sisal? What do you think of it?" These
were the questions addressed to the writer almost before he had landed
in the Bahamas. The object of the writer's visit to the "land of the
pink pearl" was to make a collection of its plants and animals; but,
during the pleasant six months occupied in so doing, he had many
opportunities of observing the cultivation of the "sisal hemp." This
industry is now in its infancy in the Bahamas, but, if the present pros-
pects are realized, it will before long bring to the islands both wealth
and prosperity. Since his return the writer has found that most of
those to whom he has spoken of sisal had at best but a vague idea of
the fiber or of the plant that produces it, so it was thought that some
notes on the subject might prove of interest.

The group of coral islands known as the Bahamas lies east of
southern Florida and north of Cuba. One of the islands. New Provi-
dence, is well known to those who, in search of health or recreation,
have been to Nassau and enjoyed its lovely winter climate. But the
"out islands," as the remaining ones are locally termed, are seldom
visited, even by those who live in Nassau. The largest of these "out
islands" is Andros, which is about the size of Long Island, New York;
there, as in all the others of the group, except New Providence, the
population is almost entirely composed of negroes, only seven white
men living on the island ; and of these, four are interested in the
production of the fiber known as sisal hemp.

The term "fiber" is used commercially to designate the material
obtained from the leaves or stems of many different plants. Hemp,
on the contrary, refers to the product of a single plant, known botani-
cally as Cannabis sativa, and belonging to the same order as our hop.
But in speaking of fibers the word "hemp" is often added, and thus
we hear of "sisal hemp," or, as it is sometimes called, "sisal grass,'*

^ Popular Science Monthly, March, 1891.

CULTIVATION OF SISAL IN THE BAHAMAS 213

or even manila. The latter term, however, is properly restricted to the
fiber obtained from a species of plantain (Musa textilis) belonging to
the same genus as the banana.

Sisal hemp, the subject of this paper, is obtained from the leaves of
some of the species and varieties of the genus Agave, one species of
which is well known in cultivation under the name of "century plant."
This genus belongs to the order Amaryllidacece, and is related to the
snow-drop, amaryllis, and narcissus; but, owing to the much greater
size of the plants, and some peculiar points of structure, it stands
prominent among its congeners. The agaves are indigenous in the New
World only, and the majority of the species are natives of Mexico,
only a few being known within the limits of the United States.

The same general appearance is presented by all, so that any one
familiar with the century plant can form a very good idea of the appear-
ance of the other species of the genus. In all, the leaves are thick and
fleshy, as they contain the supply of material which is to nourish the
great flower-stem when the plant arrives at maturity. This stem,
which is a prolongation of the trunk of the plant, shoots up from the
center of the rosette of leaves, and often attains a height of from twenty
to thirty feet. The time required to arrive at maturity varies in the
different species, and in the same species under different conditions.
The ''century plant" in its native home, Mexico, blossoms in from ten
to fifteen years, while with us it requires thirty, fifty, or in some cases,
it is said, even a hundred years to mature. During the production of
the great flower-stalk the store of nourishment in the massive leaves is
exhausted, and, after the fruit is produced, the plant withers and
dies.

The leaves of all the agaves contain what are known botanically
as the fibro-vascular bundles. In order to see these, it is only necessary
to cut off a leaf of the century plant ; as, in a thick transverse section
that has been allowed to dry slightly, the fibers will look like short
bristles projecting from the surrounding soft tissue; and in a longitu-
dinal section these bristly points are seen as threads running through
the leaf. Should the observer be the fortunate possessor of a compound .
microscope, on examining these threads he will find them composed of
exceedingly fine, elongated cells, closely connected in a bundle, and
surrounded by the much larger circular cells that compose the soft
parts of the leaf. When the outer skin and the soft tissue of the leaf

214 CULTIVATION OF SISAL IN THE BAHAMAS

are removed, the fibro-vascular bundles remain and constitute what is
commercially known as ''fiber."

While all the agaves will yield fiber of some kind, it is only in
a few that the quantity and quality of the material are such as to make
its manufacture profitable. This fact has been known for a long time
in Yucatan, the home of the sisal industry. There the natives have
from time immemorial cultivated a number of agaves, until now it is
difficult for botanists to decide whether some of them are distinct
species or only cultivated varieties.

One of the native species, known as Agave rigida, is a rather small
plant, having leaves from two to four feet long, and as many inches
wide. These are armed on the edges with dark brown spiny teeth,
and are terminated by a stout reddish brown spine. This seems to
be the plant called chelem by the natives of Yucatan, and is the one
from which the cultivated varieties are supposed to have originated.
These varieties, collectively known as henequen or jenequen, are sepa-
rately distinguished as the "yaxci, furnishing the best quality, and the
sacci, with the largest quantity of fiber; chucumci, \axger than the
last, produces coarse fiber; and babci has finer fiber, but in smaller
quantity."

Of the varieties mentioned above, only two need be considered —
the sacci and the yaxci. The former, known as Agave rigida, var.
longifolia, is distinguished from the native plant by having much
longer, spiny leaves, from four to six feet in length, and slightly differ-
ent flowers. It is extensively cultivated in Yucatan, and, as already
stated, yields the most fiber. The other variety, the yaxci, botanically
dignified by the title Agave rigida, var. sisalana, or sometimes even
elevated to the rank of a species, is one of the most valuable of the
fiber-producing agaves.

The leaves are of a dull green color, four to six feet long, as many
inches wide, and terminated by a stout, dark spine. The marginsare
commonly described as smooth, as they are without teeth, but in all
the plants examined by the writer the leaves were slightly rough on the
. edges, and in many of the young plants some of the leaves had well-
developed teeth. A full-grown plant presents a rather striking appear-
ance, bristling all over with the long, spiny-tipped leaves, thickly radi-
ating from the short cylindrical trunk, which is crowned by a sharp,
slender, cone-like bud. Indeed, a large plant makes one think of a

CULTIVATION OF SISAL IN THE BAHAMAS 215

gigantic sea-urchin. The leaves as they unfold from the bud slowly
assume a horizontal position, but remain rigid and straight, never
curving downward, as they do in the century plant.

As has been said above, when the plant arrives at maturity, and
has a sufficient store of nourishment, it sends up its flower-stem, known

Fig. 6. — Agave rigida, var. sisalana, in blossom, near Nassau, N. P.

to cultivators as the " mast " or " pole." This is from twenty to twenty-
five feet high, and about six inches in diameter near the base. On the
upper two-thirds branches are developed, converting the pole into a
huge panicle, covered with innumerable greenish yellow flowers. A
peculiarity of the sisal plant is that it seldom or never sets a seed. The
flowers fall, carrying the ovary with them, then on the ends of the

216 CULTIVATION OF SISAL IN THE BAHAMAS

branches young plants develop, so that the pole presents a rather odd
appearance, with the small plants growing out in the places usually
occupied by the flowers. When these young plants have attained a
height of from three to four inches, they fall to the ground and take
root. The old plants also reproduce themselves by means of suckers.

Fig. 7. — A Branch of the " Pole " of the Sisal Plant.

and hence, when old and neglected, are often seen surrounded by
numerous smaller ones, as in the common houseleek (Sempervivum).
Such is briefly a general description of the plant that seems destined
to occupy the capital and energies of the people of the Bahamas; for
it was this plant that was introduced there a few years ago by Sir
Henry Blake,^ then governor of the colony. Although the plants were
neglected, they throve and increased to such an extent that finally the
people looked upon them as troublesome weeds, and as such they were

^ Governor Blake is generally credited with having introduced the plants. But as
early as 1854 an agave was sent by the British vice-consul, Baldwin, from Florida to the
Bahamas. It is not unlikely that this plant was the same as those introduced by Dr.
Perrine into Florida.

CULTIVATION OF SISAL IN THE BAHAMAS 217

often destroyed. Their usefulness, however, was evidently appreci-
ated by a few ; for, as Sir Ambrose Shea, the present governor of the
Bahamas, told the writer, he was one day passing the house of a native,
when a piece of rope attracted his attention. On inquiring where he
obtained it, the negro replied that "it growed in de yard," and showed
the governor the plant, and explained the way in which the rope had
been made. Now, Sir Ambrose happened to be a native of Newfound-
land, and hence knew a good rope when he saw it ; so inquiries were
at once made, and the value of the plants was learned.

The people, however, were slow to realize the importance of the
subject, but the governor evinced great energy and enthusiasm in
keeping it before them, and when some of the fiber obtained from
old plants sold in London at the rate of fifty pounds per ton, and was
declared to be superior to that produced in Yucatan, sisal in the Baha-
mas had somewhat of a "boom," and people carefully guarded the
very plants that formerly they would have destroyed as weeds. Every-
body became enthusiastic, and sisal plantations were everywhere
started, not only by the people of the colony, but also by outsiders,
as the following facts show.

A company from St. John's, Newfoundland, has obtained a grant
of 18,000 acres of crown land at Abaco; another tract of 20,000 acres
on the same island has been allotted to a London company; 2000
acres have been taken on Andros by a gentleman from Edinburgh;
1200 are in process of cultivation on Inagua; but the largest appli-
cation has been lately made by two London companies, who together
ask for 200,000 acres. Besides the large plantations mentioned above
many small scattered areas go to swell the total. Indeed, there have
been so many demands for crown land that the governor has recently
advanced the price from one dollar and twenty-five cents to four dol-
lars per acre.

Now as to the character of the land. In Andros, which, as above
stated, is the largest of the group, and where most of the writer's time
was passed, the land is locally described by one of three terms : it is
either "coppet," "pine-yard," or "swash." The coppet, which occu-
pies, as a rule, the more elevated parts of the island, is composed of
small angiospermous trees, often only two or three inches in diameter,
and so close together as to make an almost impassable thicket. Back
of the coppet, which is mostly a fringe along the eastern coast, nearly

218 CULTIVATION OF SISAL IN THE BAHAMAS

the whole interior is one vast " pine-yard," made up of the Bahama pine
(Pinus hahamensis). The trees are generally small, and from ten to
twenty feet apart. Under them is very frequently a dense undergrowth
of a tall brake, which is often six or seven feet high, and is known by
the natives as "May-pole."

"Swash" is a very expressive term to denote the low swampy
ground, of which there are thousands of acres on the west coast. Here

1 ■

^^^m^m^^^m^.

f .

?w^4 ' - r^^^M^fe "

-'"ir-f

'' \i'f

'/| SI,

m- .

Fig. 8. — " Swash," West Side of Andros.

the soil is soft and is composed of comminuted calcareous particles;
it supports no vegetation except innumerable small mangroves (Rhi-
zophora mangle), here and there small " button- woods " (Conocarpus
erectus), a few "salt bushes" (Avicennia nitida), and in some places
palmettoes. So far as sisal cultivation is concerned, the swash is
utterly valueless; but the pine-yard and coppet are both available.
In neither of these, however, is there what we recognize here as " soil" ;
and at first it was a source of wonder to the writer that anything at all
could grow there, for the surface is very largely the bare coral rock.

CULTIVATION OF SISAL IN THE BAHAMAS 219

However, it is rarely smooth, but is rough and jagged with innumer-
able points and crevices, so as to resemble somewhat the appearance
of a well-thawed mass of snow-ice. In most places, also, there are
numerous holes, from a few inches to many feet in diameter; and it is
in these holes, cracks, and crevices that what little earth there is can be
found — still, this little seems sufficient to support the dense vegetation.
Some of the other islands — Eleuthera, for instance — have consider-

FiG. 9. — Clearing the "Pine-yard" for Sisal near Nassau, N. P.

able depth of soil; but it is when growing on the bare, rocky ground
described above that the sisal is said to produce fiber of the best
quality.

Given the land, the next step is to clear it, and the method of clear-
ing varies according to the character of the vegetation. If it is pine-
yard, a fire is started, which burns off the May-pole; the pines are
then cut down, and either made into charcoal or laid in rows across the
fields and allowed to decay; if coppet, the trees and shrubs are cut
down with axes or cutlasses, according to their size, and the brush
is then burned.

220 CULTIVATION OF SISAL IN THE BAHAMAS

While his land is being cleared, the planter should be getting his
plants ready. As usually obtained, they are fresh from the "pole,"
and only from one to four inches in height. These are too small to
put out in the fields, so they are set out in beds of cave earth until they
get to be eight or ten inches high. When taken from these nurseries
their rootlets are carefully trimmed off, and they are then planted
every eight or nine feet in rows that are about ten feet apart. Thus
an acre of ground usually contains from five to six hundred plants.
In order to facilitate carrying the leaves out of the field, the latter is
divided by roads into sections of about one hundred acres each.

After planting, it is not very long before the fields will have to be
weeded, and this process is said to be necessary about twice a year,
until the sisal plants attain a height of three or four feet, when weeding
is no longer needed. The most troublesome enemy of the planter, in
the way of weeds, is the "May-pole," as it grows very rapidly, but the
roots are said to die after the third cutting. In about four years the
sisal plant produces what are called "ripe leaves" — that is, leaves that
are horizontal and large enough to cut. The cares of the cultivator
are now about over, and all he has to do is to cut off the leaves as fast
as they mature, and manufacture his fiber.

The cultivation of sisal is of such recent introduction into the Ba-
hamas that as yet none of the large plantations have begun to pro-
duce to any extent; so for a description of the next stages we will turn
to Yucatan, where, as has been said, the industry has been carried on
from time immemorial. There the men cut the leaves off close to the
trunk, and lay them tip to butt in bundles of fifty, when they are carted
to the machines. The cutting of thirty bundles, or fifteen hundred
leaves, is considered a good day's work. In order to save the cost of
transportation, as the leaves yield but about five per cent of fiber,
there is usually a machine to every one hundred acres. The machine
now in use consists of a horizontal wheel, on the face of which brass
strips are transversely placed, forming dull knives. The leaf is intro-
duced so as to bring one side in contact with the revolving wheel,
which is run by a small engine. A brake then presses the leaf against
the scrapers, while the butt is firmly held by a pair of pincers. The
scrapers remove the outer surface and some of the soft tissue; then the
leaf is taken out and turned, and the other side undergoes the same
operation, until only the fibers are left. These are then shaken out and

CULTIVATION OF SISAL IN THE BAHAMAS 221

hung in the sun for a few hours to dry. The result is a rather coarse
fiber of considerable strength. The finest quality is nearly white,
while the inferior grades are yellowish in color. In order to produce
the best quality of fiber, the leaves must be cleaned as soon as possible
after being cut; otherwise the fiber is apt to be spotted.

It may be well to state here that the cultivation of sisal is also being
tried in Bermuda, Trinidad, and Jamaica, but on a much smaller

Fig. io. — The House of a Sisal Planter, Andros.

scale than in the Bahamas. There, as already stated, large tracts of
land have been bought from the government for the sole purpose of
producing the sisal hemp. The price is now four dollars an acre, and
two acres are said to produce one ton of fiber. Wages for men vary
from thirty-six to sixty cents per day, according to the season and
locality, as most of the negroes are spongers, and at certain times of
the year labor is not easy to obtain. Women, however, are largely
employed in the planting and weeding, and receive on the average
twenty-five cents a day. These are the data on which it is stated that

222 CULTIVATION OF SISAL IN THE BAHAMAS

a ton of fiber can be produced for $50. As the price of the fiber is
now from. $120 to $130 a ton, and has been as high as $200, these
figures look attractive.

But it may well be asked, " How about the quantity of fiber now on
the market, and will the market stand the enormous increase, that the
yield of the BahamasVill give ? " That is, of course, the very point on
which the question of profit or loss will turn. The writer has been told,
by one who is well acquainted with the fiber market, that if the sisal
hemp could be sold for four and a half or five and a half cents per pound,
in a few years the consumption would be doubled ; for, when the price
reaches nine or ten cents a pound, the use of the fiber for many pur-
poses is abandoned, and is replaced by some cheaper material, as
jute.

One of the principal obstacles in the way of cheaper fiber is the
need of a good machine, as the one now in use is a crude affair, requir-
ing the attendance of two men and a boy besides the engineer, and
producing but a small quantity of fiber daily. Although much skill
and money have already been spent in attempting to invent a better
machine, as yet all efforts have been unsuccessful; but, as inventors
and mechanics are still at work, and as the recent ''sisal boom" in the
Bahamas will increase the demand, there is little doubt but that here,
as in so many other cases, necessity will prove the mother of invention.
When the fiber can be cheaply produced in large quantities, there is
little doubt but that increased uses will be found for it, and that the
demand will equal the supply.

In 1887 Yucatan exported crude fiber valued at over $3,000,000,
besides $37,862 in rope and $43,891 in hammocks. About eighty-four
per cent of the crude fiber and fifty per cent of the hammocks came
to the United States; most of the remaining fiber went to England,
Germany, and France, while Spain took the rest of the hammocks
and all the rope. In 1889 the import of sisal hemp into the United
States was between $6,000,000 and $7,000,000, about 50,000 tons, on
which a duty of $15 a ton was paid.^

Now it may be asked: "Why cannot the United States produce
sisal too ? Is no portion of our vast territory suitable for this crop ? "
As we shall see, some one did ask that question over fifty years ago.
It is not generally known that in 1827 the Treasury Department issued

^ The duty has since been removed.

CULTIVATION OF SISAL IN THE BAHAMAS 223

a circular to some of the American consuls, requesting them to collect
and preserve seeds and specimens of such plants in their districts as
were " useful as food for man or the domestic animals, or for purposes
connected with the manufactures or any of the useful arts." The
American consul at Campeche, Dr. Henry Perrine, responded to this
call with energy and enthusiasm, and soon introduced into Congress
" a bill to encourage the introduction and promote the cultivation of
tropical plants in Florida, and conveying to Dr. Perrine and his asso-
ciates a township of land, on condition that every section should be
forfeited if at least one-fourth thereof should not be occupied and
successfully cultivated in tropical or other plants within five years."
These hard conditions were accepted by Dr. Perrine, and in one of his
letters to Congress he calls attention to the sisal plant, and says,
"He repeats his unbroken conviction that its introduction will make
an era of as great importance to the agricultural prosperity of our con-
federation as the invention of the cotton-gin."

For nearly ten years he labored, sending to Florida plants and seeds,
and endeavoring to obtain his township of land, desiring ''no more
honor than the power of passing the brief term of his painful existence
amid the privations and exposure incident to a chief pioneer in the
planting and population of tropical Florida." He finally succeeded
in establishing a sisal plantation on Indian Key. Unfortunately,
Dr. Perrine was not permitted to see the result of his labors, for, during
the Seminole War, the Indians set fire to his buildings, and he himself
fell a victim to their merciless attack. With the death of Dr. Perrine
ended the cultivation of the plants he had introduced ; but one of them,
that he named Agave sisalana, remained, became naturalized, and is
now flourishing on some of the Florida Keys, where the young plants
are now being gathered and carried to the Bahamas.

Thus we see that the plants are growing within our borders, and it
is only necessary to determine the quality of their fiber; for, although
the plants are the same species as those now cultivated in Yucatan and
the Bahamas, the quality of the fiber may not be as good, and yet on
the other hand it may be better. For instance, it is said that the Ba-
hama fiber is superior to that produced in Yucatan ; so why may not
the "Florida fiber" of the future surpass that of the Bahamas? In
order to determine its value it is only necessary to prepare it by hand
from the plants now growing in Florida and compare it with the article

224 CULTIVATION OF SISAL IN THE BAHAMAS

now on the market. The subject is being investigated by the Depart-
ment of Agriculture, and a report may be looked for in the near future.
It may be said in conclusion that, as a crop, sisal has much to rec-
ommend it. It grows best on barren, rocky land that is useless for
other agricultural purposes. Drought affects it but little, if at all, as
the writer can testify from his own observation. The yield is not con-
fined to any one season, but is continual; hence the employment of
labor is constant, and the planter can estimate closely what the yield
will be for a given time. The old plants are easily replaced by the
suckers that have been previously cut off and kept for this purpose.
These advantages are shared by all the cultivators of sisal; but, in
addition, the planter in Florida will have at his door a market that now
absorbs eight-four per cent of all the fiber produced. He will not only
bring into use land now almost worthless, but will probably make for
himself a fortune and introduce a new industry into the United States.

MISCELLANEOUS PAPERS

BY
JOHN I. NORTHROP

THE ERUPTION OF KRAKATOA IN 1883^
By John I. Northrop

Every one has heard or read of the effects of a volcanic eruption,
but the cause of the outbreak is more frequently wondered at than
understood.

The old Greeks and Romans were content to consider the volca-
noes of which they had any knowledge as the property of one of their
gods or the abiding place of departed spirits ; and for a long time super-
stition and fear prevented accurate observation of volcanic action.
About a hundred years ago the illustrious Italian naturalist, Spallan-
zani, published the results of his study of the volcanoes of his native
land, and since that time many workers have entered the field, and our
knowledge has steadily increased, until a science has been created and
called Vulcanology.

The volcanoes of the world, with two exceptions, have been found
to form four irregular lines upon its surface, thus indicating the exist-
ence of as many great fissures in the so-called crust of the earth. The
exceptions are the volcanoes of the Sandwich Islands and those of the
Thian Shan range in the central part of Asia. The lines of volcanic
action, as a rule, follow the coasts of the continents, and with the
exceptions already noticed, the volcanoes are situated in close prox-
imity to the bodies of water they border. This fact has led vulcanolo-
gists to look to the oceans for an explanation of volcanic eruptions.

The line separating fusion from solution is not a sharp one, for it
has been shown by adding niter to a given quantity of water, that when
the cold solution becomes saturated, it is only necessary to apply heat
to dissolve still more of the salt. By increasing the heat and the addi-
tion of niter, a point may be reached where the mixture can no longer
be called a solution, but must be considered a fusion. It has also been
proved that rocks containing a certain percentage of water are more

^ School of Mines Quarterly, Vol. X, No. 2, January, 1889, pp. 1-15.

225

226 THE ERUPTION OF KRAKATOA IN 1883

fusible than those that contain a smaller proportion, and it is well
known that among the minerals some, as the zeolites, are fusible,
while others, containing no water, are practically infusible.

The primary cause of a volcanic outbreak is the heat that is present
below the surface of the earth, but it is the slow, very slow, percolation
of a small quantity of water through the rock that is supposed to aid
in the fusion. During this process gases are produced that gradually
increase in pressure, until at last force enough is generated to overcome
the weight above, and an eruption is the result. It has been suggested
that a sudden inrush of the sea through a fissure to the molten lava
would give rise to an evolution of steam sufficient to cause an eruption,
but Professor Judd claims that the effect upon the lava would be to
form a crust of chilled material under the water, which would confine
the volatile products generated below, and only when the pressure of
these gases became greater than the load above would an outbreak
occur.

Most of the material ejected from a volcano falls close to the vent,
thus forming a cone that continues to increase in height, until an erup-
tion of exceptional violence entirely removes the upper portion, leaving
a ring or crater that Darwin has aptly called the "basal wreck."
Within this crater a new cone may be built up, and the former outline
of the mountain be restored, to be once more destroyed in its turn.

In the Strait of Sunda, between the islands of Java and Sumatra,
are portions of an old crater ring, which, rising above the water, form
the islands of Krakatoa, Verlaten, Lang, and Polish Hat. One of the
grandest lines of volcanic activity on the earth runs through these
islands, and is continued eastward through the islands of Bali, Sum-
bawa, Flores, and Timor. There is evidence that a second fissure
crosses this main one at nearly right angles, and at the intersection of
these two lines Krakatoa is situated. Notwithstanding the important
position it occupied, until recently this little island received but a
small share of the attentibn that geologists have given to volcanoes in
general. But this is not to be wondered at when it is remembered that
Java contains "no fewer than forty-nine volcanic mountains, some of
which rise to a height of 12,000 feet above the sea level," and a few are
in a state of almost constant activity. In recent geological times
Krakatoa, though now insignificant, rivaled these mountains in size,
for if the conclusions that geologists have drawn are correct, the present

THE ERUPTION OF KRAKATOA IN 1883 227

islands of Krakatoa, Verlaten, Lang, and Polish Hat are but the re-
mains of the '' basal wreck" of a volcano that rose 10,000 feet above the
sea, with a circumference, at its base, of twenty-five miles. At some
unknown period an eruption took place that removed the central mass
of this volcanic mountain, leaving a crater, about four miles in cir-
cumference, the highest points of which were but a few hundred feet
above the surrounding water.

After a series of quiet eruptions that gradually filled up this old
crater, a lateral or parasitic outburst seems to have taken place at its

lou^

6\ d xT iv-»^WWw

VVVW O^Q^ w> VV C^ Vt . ^^'"^^"^^

Fig. I . — Map of the Islands of the Krakatoa Group before the Eruption of

August, 1883.

From the "English Krakatoa."

southern edge. This activity must have continued for a long time, as
it resulted in the formation of a cone about 2600 feet in height. It is
this cone "that was called by the natives the peak of Rakata, which,
in the old Kawi or Javanese language, signifies a crab. The name,
under the Dutch form, Krakatau, the Portuguese Krakatao, and the
English Krakatoa, has been extended to the whole island upon which
the striking cinder cone stood."

From the meager accounts that have been preserved, it seems that
in May, 1680, an eruption occurred at some point on the island, and it
is probable that Perboewatan was in activity, as fresh lava streams
have been seen on its sides by several later authors.

The relative positions of Krakatoa and the adjacent islands are
shown in Figure i . The broken line indicates the form of the old crater,
of which the islands of Verlaten and Lang are portions rising above the

228 THE ERUPTION OF KRAKATOA IN 1883

. strait. The cones within the ring were formed by the quiet eruptions
that filled up the crater. The largest cone to the southwest is the peak
of Rakata. During the late eruption almost all of the island Krakatoa,
north of the line drawn through this peak, was removed. The increase
of size of Verlaten Island is indicated by the line surrounding it. Lang
Island was also slightly added to at the north.

Figure 2 is a section through the length of] the island of Krakatoa
and shows its geological position. A thin bed of postertiary age (a)
rests on the tertiary rocks below, and above it stood the old volcano,
whose supposed form is indicated by the broken line. The lateral
cone of Rakata is well shown, and also the position of the two small
cones — Danan and Perboewatan.

After the eruption of Perboewatan in 1680, for nearly two centuries
Krakatoa remained scarcely noticed amid the grander peaks that sur-
rounded it. But during this time the unseen forces below had been
gathering strength, as if to resent this seeming neglect, and the first
intimations of what followed were given by earthquakes, of frequent
occurrence, that were felt as far distant as North Australia.

On May 20, 1883, ''sounds like the firing of artillery were heard
at Batavia and Buitenzorg in Java, nearly 100 English miles from Kra-
katoa." On the twenty-sixth of the same month a party was organized
at Batavia and proceeded in a steam launch to the island. It was then
seen that the cone of Perboewatan was again in activity. The party
succeeded in taking a photograph of the volcano and returned to
Batavia. The accounts of the subsequent disturbances are rather
indefinite. According to Professor Judd, ''the eruption from Perboe-
watan seems to have had sufficient force to carry the volcanic dust to
various points along the shores of Java and Sumatra. After this
sudden outburst there was a rapid and marked decline in violence, and
then a gradual increase till June 24, when a second crater had opened
in the center of the island. The eruptive force still increasing, a third
crater made its appearance, and innumerable smaller vents were
originated all over the surface of the filled-up crater of the great vol-
cano. From this time the activity seemed constantly to increase till
its grand culmination on the 27th of August."

On that afternoon and through the succeeding night the forces
Krakatoa had so long imprisoned broke loose with a violence un-
equaled in the records of volcanic activity, and caused the scientific

bO ^
o f-

229

230 THE ERUPTION OF KRAKATOA IN 1883

world to give this little island the attention for which it had waited so
long.

Shortly after the eruption the Dutch government caused a careful
survey of the vicinity to be made, and commissioned Mr. R. D. M.
Verbeek, a mining engineer of Batavia, to investigate the cause and
effects of the outbreak.

Later, the Royal Society of London took the matter in hand, and
on January 17, 1884, the council passed the following resolution:
"That a committee, to consist of Sir F. Evans, Professor Judd, Mr.
Norman Lockyer, Mr. R. H. Scott, General Strachey, and Mr. G. J.
Symons, with power to add to their number, be appointed to collect
the various accounts of the volcanic eruption at Krakatoa, and atten-
dant phenomena, in such form as shall best provide for their preserva-
tion and promote their usefulness." The committee inserted a letter
in the Times and other periodicals, inviting communications relating
to the eruption. Later, Dr. Geikie and Professor Bonney were added
to the committee. The work of discussing the mass of data collected
was commenced about the end of November, 1884. In the spring of
1887 the manuscript was completed, and has since been printed in a
quarto volume of four hundred and ninety-four pages. The results
of Mr. Verbeek's investigations were published in Dutch in 1885-1886,
and afterward translated into French.

It was suggested to me that a description of the eruption might be
of interest to the readers of the Quarterly. Most of this article is
based on the English work, as I have been unable to obtain the first
volume of Mr. Verbeek's account.

During the night of August 27, 1883, the British ship, Charles
Bat, being prevented by the intense darkness from continuing her
voyage, beat about within a dozen miles of the east coast of Krakatoa.
This vessel and two others in the vicinity escaped destruction. About
40 miles from the scene of the eruption were four other vessels, three
being at the eastern and one at the western entrance of the strait;
while many were at distances varying from 75 to 1000 miles from the
volcano. The account of what happened during that eventful night
has been compiled from the data furnished by the officers and passen-
gers of these vessels, and the reports collected with so much care by
Mr. Verbeek.

The northern portion of the island of Krakatoa was almost entirely

232 THE ERUPTION OF KRAKATOA IN 1883

removed, and where the land had been 300 feet high a hollow
was formed over a thousanti feet below the sea level. The northern
part of the peak of Rakata was blown away, exposing a magnificent
section, Figure 4, that has enabled geologists to study the internal
structure of the volcano. The island Polish Hat disappeared entirely,
and Lang and Verlaten were increased in size; the former slightly, the
latter to about three times its former area; while a considerable addi-
tion was made to the height of both. The materials thrown from the
volcano seem to have covered the sea bottom, within a radius of 10

Fig, 5. — Outline of the Crater of Krakatoa as it is at the present time.

The broken line indicates the portions blown away during the eruption of August, 1883,
and the changes in form of the flanks of the mountains by the fall of ejected materials
upon them. From the " English Krakatoa."

or 12 miles, to a depth of from 12 to 60 feet. There is some doubt on
this subject, however, as the waters surrounding the island had never
been carefully sounded, and no accurate survey had been made. There
is good reason to believe that the floor of the ocean was raised during
the eruption, causing the great sea waves that will be referred to below.

Along a line about nine miles in length, extending westward from
the great crater, a depression seems to have taken place which "was
not improbably due to the opening of a fissure on the flanks of the
submerged cone." During the eruption two small islands were formed,
but both were soon reduced to shoals by the action of the waves. These
islands are supposed to have been caused by parasitic cones that had
arisen upon the northern flank of Rakata.

The watery vapor thrown into the air reached a height variously
estimated from 12 J to 17 and possibly 23 miles. On reaching the limit
of its elevation, the mass spread itself and gave rise to the "pine-tree"
appearance so often seen during the eruptions of Vesuvius. The sur-
rounding country was enveloped in darkness for nearly twenty-four
hours. The electrical phenomena accompanying the explosion were
on a grand scale; the lightning darted through the dark clouds "like
huge serpents," and "balls of fire" rested on the masts and yardarms
of vessels 40 miles away.

THE ERUPTION OF KRAKATOA IN 1883 233

Volcanic dust rose to a great height and, floating away, gave rise
to the wonderful optical phenomena noted throughout the civilized
world. Some of the air waves generated caused the cracking of walls
100 miles away, and others less violent, though of greater dimensions,
travelled nearly four times round the earth. The terrible explosions
gave rise to vast sea waves that stranded vessels, and devastated the
shores of Sumatra and Java, causing the death of over 36,000 human
beings. These explosions formed a notable feature of the eruption,
and were probably due to the peculiar situation of the volcano. Owing
to the proximity of the ocean, the water, at times, must have rushed in
on the molten material. This would at first cause an evolution of
steam and then a diminution of activity in the volcano, thus causing a
"check," only to be followed by an outburst or "rally" of greater
violence. In this way the alternations of the "check and rally" are
supposed to have caused the unprecedented explosions.

The eruption continued until the 28th or 29th of August, with
perhaps a slight outbreak on the loth of October of the same year.
With this last sign of activity the subterranean forces seem to have
exhausted their energy, and Krakatoa once more relapsed into a period
of rest. Such is a general view of the eruption. We will now examine
a few points a little more in detail.

Materials Ejected

The materials ejected from Krakatoa were lava, pumice, and vol-
canic dust. The lava presents two distinct forms, characterized as
porphyritic obsidian and porphyritic pitchstone. Both contain the
same crystalline elements which constitute about 10 per cent of the
rock; these crystalline elements are feldspar, pyroxenes, and magnetite
scattered irregularly throughout the ground mass. In the obsidian
the crystals of feldspar have perfectly sharp outlines, while in the pitch-
stone they are very much corroded, probably by the action of the fluid
material which once surrounded them. The most striking difference
between these two rocks is shown when they are placed in the flame
of a blast-jet. The pitchstone is found to decrepitate and to undergo
fusion with the greatest difficulty; the obsidian, on the other hand, fuses
with comparative ease, and during the process forms "cauliflower-like
masses" that are identical with the pumice thrown from the volcano.

234 THE ERUPTION OF KRAKATOA IN 1883

The loss suffered by the pitchstone during fusion is nil, while the ob-
sidian loses from i, to 5 or 6 per cent of its weight. There is no
doubt that the pumice ejected in such enormous quantities was formed
by the expansion of the volatile gases contained in the obsidian. The
condition in which the water and gases existed in the rock is a problem
worthy of study.

The pumice contains the same crystals as the obsidian, and also
little pieces of a black glass that have been shown to be identical with
the pitchstone already referred to. Mr. Verbeek has calculated that
the pumice and dust form 95 per cent of all the materials ejected.
Some of the smaller pieces of pumice were thrown to a great height,
and were carried by the wind some distance from the scene of eruption ;
the greater quantity, however, fell close to the volcano,, covering the
sea, in places, to such a depth as to effectually hinder the progress of
vessels.

As the larger pieces of pumice fell back into the crater, they came
in collision with others that were being thrown out, and in this way the
pieces were broken, the finer particles forming the volcanic dust. The
dust fell in great quantities in the vicinity, and samples have also been
collected from many localities, some as far as a thousand miles from
the island. Its exceeding fineness is shown by the fact that it pene-
trated everywhere, no box or watch being sufficiently tight to exclude
it. The particles composing the dust have been separated and ana-
lyzed, and their optical properties investigated. The appearance and
composition of the dust varies considerably ; that found in the vicinity
is heavier, darker in color, and more basic in composition than that
gathered at a greater distance from Krakatoa. This difference is due
to the heavy particles of magnetite pyroxene, and feldspar it contains.

The quantity of dust thrown into the air is not known, but it was
probably much less than that formed at some previous eruptions. The
work of trituration, however, continued about three months, and the
last paroxysmal explosions of Krakatoa were altogether of exceptional
violence and carried the finer particles high into the rarer regions of the
atmosphere. From the data supplied by the vessels on which the dust
fell, it is found to have covered an area of 1,100,000 square miles.
From the depth of the material it is calculated that the quantity of dust
was equal in bulk to 3^ cubic miles, while the entire mass of ejected
matter equalled about ^\ cubic miles. These figures, however, are
estimates and probably only approximate the actual quantities.

THE ERUPTION OF KRAKATOA IN 1883 235

Air Waves

The air waves generated by the eruption were of three kinds:
first, those sufficiently rapid to give rise to sound ; second, larger waves
that caused the breaking of windows and the cracking of walls, a
hundred miles away, at Batavia; and third, those of still greater dimen-
sions that travelled several times around the earth.

The sounds caused by the explosions were heard at Ceylon to the
northwest; Perth and other stations in Australia to the southeast, and
at Manila to the northeast. These places are situated close to a circle
drawn from Krakatoa as a center with a radius of 30 degrees. To
the west the sounds were carried by the trade-winds a much greater
distance, and were heard at the island of Rodriguez, nearly 3000
English miles from the scene of eruption. This is interesting not only
as being the most remote point at which the reports of the explosions
were noticed, but as "the only instance on record of sounds having
been heard at anything like so great a distance from the place of their
origin." The area over which the sounds were recorded is about one-
thirteenth of the earth's surface.

The detonations in the vicinity of the volcano, although very vio-
lent, were not as loud as might be inferred from the distance at which
they were heard, for the great rain of pumice and the heavy cloud of
dust seem to have acted as a curtain through which the sound waves
were unable to penetrate. At some places in the immediate neighbor-
hood of the volcano sounds ceased to be heard after ten o'clock on the
morning of August 27; although it is known that the explosions
continued with great intensity after that time.

The great air wave that owed its origin to one of the last paroxysmal
outbursts of Krakatoa was recorded on the barograms at the observa-
tories scattered over the earth. Theoretically an air wave caused by a
sudden explosion would spread in a circle around the earth till it
reached a point 90 degrees from its origin; it would then travel onward,
contracting as it advanced, till at the antipodes it would come to a
focus and thence be reflected back to its starting-point, from whence it
would once more travel to the antipodes and so on, till it gradually
became lost. The barograms that have been collected show that the
great wave passed seven times round the earth, four times from
Krakatoa to the antipodes, and three times from the antipodes to

236 THE ERUPTION OF KRAKATOA IN 1883

Krakatoa, the barometers failing to record the fourth return of the
wave.

The velocity of the wave varied from 674 to 726 English miles per
hour, being very nearly equal to the velocity of sound in air at 50°
Fahrenheit, which is 757 miles. There is no direct evidence, however,
that the great explosion producing the wave was accompanied by
sounds sufficiently loud to be heard at any great distance. The time
occupied by the wave passing around the earth was about 36 hours.
The wave moving with the rotation of the earth traveled with a slightly
greater velocity than that going in the opposite direction, viz. about
28 English miles per hour. This difference is supposed to be due to
the winds that blow in a general westerly direction on the path followed
by the wave. From the velocity of the wave it has been attempted to
deduce the time of the great explosion, and the result agrees very closely
with the record made by the gasometer at Batavia; for, when the wave
passed over that place, the gasometer leaped from its well, and left a
plain, though not very delicate, mark of the time. As thus determined
the explosion occurred at 9 hours 58 minutes, local time, on August
27.

Sea Waves

No sea waves were caused by the eruption of Krakatoa during
May, but on the morning of August 27 several small ones inun-
dated the shores of Sumatra and Java, partially destroying the towns
and villages, and causing the death of many of the inhabitants. The
number and size of the waves that followed are not known, for the
people had fled terror-stricken to the hills, and, surrounded by
darkness, were but little fitted to note the changes about them. The
most destructive waves followed the great explosions of the volcano
some time after ten o'clock on the morning of August 27. From the
data collected by Mr. Verbeek it has been calculated that the largest
of the waves was about 50 feet high before it reached the shore,
although the traces it left on the land show that it rose 70 feet above the
sea level, and its height has been estimated by some as great as 125
feet. The smaller waves are supposed to have been due to the sudden
evolution of steam, caused by the contact of the sea and the molten
lava, and by the fall of huge portions of the volcano. The origin of
the great wave is an unsettled question. It has been assumed that

THE ERUPTION OF KRAKATOA IN 1883 237

during the eruption the volcano was raised bodily into the air and fell
into the water, thus causing the wave; but this does not seem probable,
as the mean period between crest and crest of the waves that followed
the great one was two hours, and, further, there is no direct evidence
supporting this view. It. has also been suggested that if an enormous
chasm had opened in the floor of the ocean the inward rush of the water
would be sufficient to create the wave, — but in this case the fissure
would have to be large enough to allow the water to flow into it for an
hour with a rapidity sufficient to cause a great decrease in the level of
the sea, and the first wave reaching the shore would be a negative one,
which it was not. It is probable that the great wave was caused by an
upheaval of the sea bottom, as it is known that the depth of water in the
strait was decreased during the eruption, and this may have been due
to an upheaval, the fall of ejected material, or to both causes. One of
the waves, though not as high as the others, was longer, and probably
traveled to Cape Horn, 7520 miles away, and possibly even as far as
the English Channel, a distance of 11,040 miles. At such distances the
height of the waves becomes very small, and hence their identification
is rendered difficult.

The study of the velocity of seismic waves is of recent date. The
first observations recorded were made by Professor W. P. Trowbridge
at San Diego, CaL, in 1855. The waves then noticed were found
to have originated near Simoda, Japan, and were due to a submarine
earthquake. Since then seismic waves have been found to be of fre-
quent occurrence, although it is seldom that they can be traced to their
starting-point. This is a matter of some interest, for the speed of the
waves bears a constant relation to the depth of the water through
which they travel. The velocity of those caused by the eruption of
Krakatoa has been carefully calculated, for in this case the time and the
place of origin were known. Some doubt is thrown on the result, how-
ever, as the same wave could not be positively identified at different
places, owing to the rather inclosed position of Krakatoa, and hence
the irregular path followed by the undulations; again, some of the
tide gauges were situated up rivers and in estuaries where the direct
effects of the wave would not be felt. It is quite certain, however, that
the velocity of the waves from Krakatoa was less than that demanded
by theory, as has also been observed in all previous cases.

238 THE ERUPTION OF KRAKATOA IN 1883

Optical Phenomena

No effect connected with the Krakatoa eruption has been so widely
observed and speculated upon as the wonderful appearance the sky
presented during the two years that followed the outbreak. Many
theories have been offered to explain the phenomena, and while most
of these agree that the dust formed in such enormous quantities during
the eruption was the primary cause, some contend that the two were in
no way connected. It is impossible to give here even a short account of
these numerous theories, or the reasons why many have been rejected.

Meager accounts exist of similar phenomena accompanying pre-
vious great eruptions. During the 250 years, from 1500 to 1750 inclu-
sive, there were 65 in which eruptions were recorded, nine of which were
of exceptional magnitude, " and in three of these latter years Denmark
was affected by appearances, two of which certainly were of the char-
acter of the afterglows, while the third is doubtful."

From 1750 to 1886 the number of "years of eruption" was 90;
25 of the outbreaks being of great magnitude, and of these 16 appear
to have been followed by " unusual redness in the sky, or brightness
after sunset." Thus it will be seen that in the past like causes have
produced like effects, and the lack of information on the subject is
probably the fault of observers rather than the want of opportunity for
observation. The optical effects of the eruption of Krakatoa, however,
were exceptional, and had previous phenomena been so marked and so
widely spread over the earth, they would have received more attention
than the records imply. The effects of the outburst at Krakatoa were
the unusual twilight glows, the blue, green, and other appearances of
the sun, the sky haze, and the large corona round the sun and moon
— all these phenomena were noticed first in the vicinity of Krakatoa.

The dust, after being ejected from the volcano, was carried west-
ward around the earth at the equator, certainly twice, and "perhaps
continuously." From this band. the particles appear to have spread
north and south, and so the phenomena were witnessed throughout
the world. The twilight glows traveled westward at the rate of about
72 miles an hour, while the colored suns moved at the rate of about 70
miles. The unusual twilight glows are supposed to have been caused
by the reflection of the sun's rays from the glassy particles of the pow-
dered pumice that were suspended about 16 miles above the earth.

THE ERUPTION OF KRAKATOA IN 1883 239

The colored suns were probably due to the absorption of certain rays
of light by the volcanic dust in the atmosphere. The large corona
around the sun was first recorded by Mr. Bishop at Honolulu, on Sep-
tember 5, 1883, and since that time it has generally been known as
''Bishop's ring" ; the halo around the moon was noticed in the tropics
during the same month. These effects are referred to the lighter
particles of dust in the atmosphere, for in April, 1884, when most of
the coarser material had settled by gravitation, and the twilight glows
began to fade, the corona brightened, then gradually became fainter,
and was last observed in June, 1886, after which the sky resumed its
normal appearance.

In conclusion, it may be said that the works on Krakatoa, referred
to above, contain many charts, maps, and barograms, and in short
form a complete record of all the phenomena caused by the eruption.
These works are more valuable than if they contained only the results
of the reports collected, for, by publishing the sources of information,
opportunity is given to any one interested in the subject to study these,
and thereby form his own conclusions. When other eruptions shall
have been treated in the same manner, more light will be shed on some
of the effects now slightly obscured, as the cause of the colors observed
in the sky and the origin of the great sea waves. Although eruptions
have occurred in the past, exceeding that of Krakatoa both in duration
and in the quantity of material ejected, such a complete history of one
has never been written till now. This is due in part to the spread of
civilization over the earth, and hence the establishment of places of
scientific observation. As the number of such observatories will con-
tinue to increase, it may be confidently predicted that the phenom-
ena accompanying violent volcanic outbreaks in the future will be
more closely observed and more widely recorded.

PLANT NOTES FROM TEMISCOUATA
COUNTY, CANADA^

By John I. Northrop

The parish of Notre Dame du Lac lies on the western shore of
Lake Temiscouata, Temiscouata County, Canada, about forty-eight
miles from Riviere du Loup and thirty-one miles from Edmundston,
N.B.

The surrounding country is for the most part rolling hills covered
with the usual second growth of spruce, fir, larch, and birch. There
are also groves of sugar maple and yellow birch, and on the lowland
near the lake the balsam poplar, Populus balsamifera, grows luxuriantly.
Populus tremuloides, Acer rubrum,A. spicatum, Prunus Pennsylvani-
cum, and Rhus typhina are common treees of the region.

The lake is about thirty miles long, and varies from three-quarters
of a mile to two miles in width. The shore is in most places of slate,
forming rocky points running into the lake, sometimes so abruptly as
to make passing exceedingly difficult. In the crevices of the slate grow
Lobelia Kalmiiy Campanula rotundifolia, and Parnassia Caroliniana.
Near the bank we find Potentillafruticosa, Spircsa salicifolia, and My-
rica gale.

Farther up Alnus viridis, Cornus stolonifera, and Viburnum Opulus
form thickets guarded by the ever present Joe Pye Weed, Eupatorium
purpureum. Here and there along the bank the berries of Pyrus
sambucifolia add their bright color to the scene.

Along the roadside, where not shaded by trees, Artemisia vulgaris
grows in undisturbed luxuriance, in company with the wild raspberry,
Rubus strigosuSy and Sambucus racemosa.

In the fields one misses the ubiquitous Chrysanthemum Leucanthe-
mum, but its place is well filled by Anaphalis margaritacea and the
Canada thistle. In many places the banks of the roadside are covered
with the bunch berry, Cornus Canadensis, which the French Canadians

* Contributions from Herbarium of Columbia Coll. No. 3. Bull. Torrey Botanical
Club, November, 1887, pp. 230-238.

240

PLANT NOTES FROM TEMISCOUATA COUNTY 241

aptly called "La rouge." Driving back from the lake, five or six ranges
of hills are crossed in as many miles. In the valleys are lakes bordered
with cedar swamps of Thuja occidentalis. The lakes are shallow and
have a muddy bottom. Cassandra calyculata, Kalmia angustifolia,
Ledum latifolium, are common on the borders.

The road from Notre Dame du Lac to Riviere du Loup runs
partly through scattered farms and partly through woodland. The
land, where cleared and neglected, is covered with a dense growth of
Pteris aquilina and Epilobium spicatum.

From Riviere du Loup I drove to Cacouna, a distance of four miles
down the St. Lawrence. The beach at Cacouna is rocky in some places
and at others sandy. On the sand Cakile Americana, Lathyrus mariti-
mus, Mertensia maritima, and Arenaria peploides were collected. On
the rocky points Plantago maritima was found abundant.

The plants of the following list were mostly collected during the
month of August, at Notre Dame du Lac. A few are from Grand
Falls, N.B., where a stay of a few days was made.

Dr. N. L. Britton has kindly annotated the list, and many thanks
are due him for his assistance and for suggestions in regard to nomen-
clature.

CATALOGUE

RANUNCULACE^

Thalictrum polygamum Muhl. Notre Dame du Lac.

Anemone muUifida Poir. Grand Falls.

Anemone dichotoma L. Notre Dame du Lac.

Anemone Virginiana L. Same locality.

Ranunculus aquatilis L., var. trichophyllus Chaix. Same locality.

Ranunculus acris L. Same locality.

Ranunculus Pennsylvanicus L. Same locality.

Actcea spicata L., var. rubra Michx. Same locality.

SARRACENIACE^

Sarracenia purpurea L. Notre Dame du Lac.

CRUCIFER^

Raphanus sativus L. Cacouna.
Cakile Americana Nutt. Same locality.

242 PLANT NOTES FROM TEMISCOUATA COUNTY

VIOLARI^

Viola hlanda Willd. Notre Dame du Lac.

CARYOPHYLLE^

Silene inflata Smith. Notre Dame du Lac.
Cerastium viscosum L. Same locality.
Stellaria media (L.) Smith. Same locality.
Arenaria peploides L. Cacouna.
Spergula arvensis L. Notre Dame du Lac.

HYPERICINE^

Hypericum mutilum K. Notre Dame du Lac.

Hypericum VirginicumL,. (Elodes Virginica Nutt) Same locality.

MALVACE^

Malva moschata L. Notre Dame du Lac.

GERANIACE^

Oxalis Aceto sella L. Grand Falls.

Impatiens hiflora Walter, 1788 (7. fulva Nutt., 1818). Notre Dame
du Lac.

ILICINEJE

Nemopanthes Canadensis (Michx.) DC. Notre Dame du Lac.

SAPINDACE^

Acer ruhrum L. Notre Dame du Lac.
Acer saccharinum Wang. Same locality.
Acer spicatum Lam. Same locality.

ANACARDIACE^

Rhus typhina L. Notre Dame du Lac.

LEGUMINOS^

Trifolium agrarium L. Notre Dame du Lac.

Trifolium pratense L. Same locality. Also a white-flowered form.

Trifolium repens L. Same locality.

PLANT NOTES FROM TEMISCOUATA COUNTY 243

Astragalus alpinus L. Grand Falls.
Vicia Cracca L. Notre Dame du Lac.
Lathyrus maritimus (L.) Bigelow. Cacouna.

ROSACEiE

Prunus Pennsylvanica L. Notre Dame du Lac.

Spircea salicifolia L. Same locality.

Rubus strigosus Michx. Grand Falls.

Ruhus triflorus Richards. Notre Dame du Lac.

Geum strictum Ait. Same locality.

Fragaria vesca L. Grand Falls.

Fragaria Virginiana Duschene. Notre Dame du Lac.

Potentilla Anserina'L. Temiscouata Lake.

Potentilla fruticosa L. Notre Dame du Lac.

Potentilla Norvegica L. Same locality.

Potentilla palustris (L.) Scop. Same locality.

Agrimonia Eupatoria L. Same locality.

Poterium Canadense L. Cacouna.

Rosa blanda Ait. Notre Dame du Lac.

Pyrus sambucifolia Cham, and Sch. Same locality.

SAXIFRAGES

Mitella nuda L. Grand Falls.

Parnassia Caroliniana Michx. Notre Dame du Lac.

Ribes lacustre Poir. Grand Falls.

DROSERACES

Drosera rotundifolia L. Notre Dame du Lac.

HALORAGEiE

Hippuris vulgaris L. Dube's Lake, Temiscouata.

ONAGRARIS

Epilobium spicatum Lam. Notre Dame du Lac.
(Enothera biennis L. Cacouna.
Circcea alpina L. Notre Dame du Lac.

244 PLANT NOTES FROM TEMISCOUATA COUNTY

UMBELLIFER^

Sanicula Marylandica L. Notre Dame du Lac.
Conioselinum hipinnatum (Walter) Britton/ Same locality.

ARALIACE^

Aralia hispida Vent. Notre Dame du Lac.
Aralia nudicaulis L. Grand Falls.
Aralia racemosa L. Same locality.

CORNACEiE

Cornus canadensis L. Grand Falls.
Cornus stolonifera Michx. Same locality.

CAPRirOLIACE^

Samhucus racemosa L. Notre Dame du Lac.
Viburnum lantanoides Michx.^ Same locality.
Viburnum Opulus L. Same locality.
Linncea borealis L. Same locality.

Diervilla trifida Moench. Same locality; also at Grand Falls, the
latter a form with leaf-margins abundantly ciliate.

RUBIACE^

Galium trifidum L., var. pusillum Gray. Notre Dame du Lac.
Galium triflorum Michx. Grand Falls.

COMPOSITE

Eupatorium ageratoides L. Notre Dame du Lac.

Eupatorium purpureum L.

Solidago bicolor L., var. concolor Torr. and Gray. Same locality.

Solidago macrophylla Pursh. Same locality.

Solidago neglecta T. and G. Grand Falls.

^Conioselinum hipinnatum (Walter, " Flor. Car.," p. 115, under Apium, 1788;
Selinum Canadense Michx., 1803). I am following Coulter and Rose in keeping up
the genus Conioselinum. — N. L. B.

2 If the V. alnifolia of Marshall is this plant, the name has priority; it. is referred
to F. lantanoides by Dr. Gray, in " Syn. Flor.," while De Candolle, " Prodr.," IV, p. 328,
refers it to V. molle Michx. — N. L. B.

PLANT NOTES FROM TEMISCOUATA COUNTY 245

Solidago serotina Ait. Notre Dame du Lac.

Erigeron annuus Pers. Same locality.

Erigeron Canadensis L. Same locality.

Erigeron hyssopifolius Michx. Grand Falls.

Erigeron strigosus Muhl. Notre Dame du Lac.

Aster diffusus Alt. Same locality.

Aster macro phyllus Li. Same locality.

Aster puniceus L. Same locality.

Aster umbellatus Mill. Same locality.

Anaphalis margaritacea (L.), Benth. and Hook. Same locality.

Achillcea Millefolium L. Same locality.

Chrysanthemum Leucanthemum L. Same locality.

Artemisia caudata L. Grand Falls.

Artemisia vulgaris L. Notre Dame du Lac.

Senecio aureus L., var. Balsamitce T. and G. Grand Falls.

Cnicus arvensis (L.) Hoffm. Notre Dame du Lac.

Arctium Lappa L. Same locality.

Hieracium Canadense Michx. Same locality.

Hieracium scahrum Michx. Same locality.

Prenanthes altissima L. Same locality.

Prenanthes serpentaria Pursh. Grand Falls.

Lactuca leucophcea (Willd.) Gray. Notre Dame du Lac.

Sonchus arvensis L. Cacouna.

CAMPANULACE^

Lobelia Kalmii L. Notre Dame du Lac.
Campanula rotundifolia L. Grand Falls.

VACCINIACE^

Vaccinium Canadense Kalm. Notre Dame du Lac.
Vaccinium Pennsylvanicum Lam. Same locality.

ERICACE^

Cassandra calyculata (L.) Don. Notre Dame du Lac.
Andromeda Polifolia L. Same locality.
Kalmia angustifolia L. Same locality.
Ledum latifolium Ait. Same locality.
Chimaphila umbellata (L.) Nutt.

246 PLANT NOTES FROM TEMISCOUATA COUNTY

MONOTROPE^

Monotropa uniflora L. Notre Dame du Lac.

OLEACE^

Fraxinus viridis Michx. A form with pubescent branchlets.

GENTIANS^

Halenia deflexa (Smith) Griseb. Notre Dame du Lac.

BORAGINE^

Mertensia maritima (L.), Gray. Cacouna.
Myosotis laxa Lehm. Notre Dame du Lac.

SCROPHULARINE^

Linaria vulgaris Mill. Notre Dame du Lac.
Chelone glabra L. Same locality.
Veronica Americana Schwein. Same locality.
Castilleia pallida Kunth. Grand Falls.
Euphrasia officinalis L. Notre Dame du Lac.
Rhinanthus Crista-Galli L. Same locality.

LENTIBULARIEiE

Utricularia vulgaris L. Dube's Lake, Temiscouata.
Utricularia intermedia Hayne. Same locality.

LABIATE

Mentha Canadensis L., var. glahrata Benth. Notre Dame du Lac.

Nepeta Cataria L. Same locality.

Brunella vulgaris L. Grand Falls and Notre Dame du Lac.

Scutellaria galericulata L. Notre Dame du Lac.

Scutellaria lateriflora L. Same locality.

Stachys palustris L. Same locality.

Galeopsis Tetrahit L. Same locality.

PLANTAGINE^

Plantago maritima L. Cacouna.

PLANT NOTES FROM TEMISCOUATA COUNTY 247

CHENOPODIACEiE

A triplex patula L., var. hastata Gray. Cacouna.

POLYGONACEiE

Polygonum aviculare L. Notre Dame du Lac.

Polygonum Hydropiper L. Same locality.

Polygonum lapathifolium Ait., var. incanum Koch. Same locality.

Rumex Aceto sella L. Grand Falls.

Rumex salicifoUus Weinm. Cacouna.

Rumex Brittanicus L. Notre Dame du Lac.

EUPHORBIACE^

Euphorbia Helioscopia L. Notre Dame du Lac.

MYRICACE^

Myrica Gale L. Notre Dame du Lac.

CUPULIFER^

Betula papyracea Ait. Grand Falls.

A Inus viridis DC. Notr e D ame du L ac .

Corylus rostrata Ait. Grand Falls.

SALICINE^

Salix discolor Muhl. Notre Dame du Lac.
Salix humilis Marsh. Same locality.
Salix lucida Muhl. Same locality.
Populus halsamifera L. Same locality.
Populus tremuloides Michx. Grand Falls.

ORCHIDEiE

Spiranthes Romanzoffiana Cham. Notre Dame du Lac.
Habenaria hyperborea R. Br. Same locality.
Habenaria psycodes (L.) Gray. Same locality.

IRIDEiE

Iris versicolor L. Notre Dame du Lac.

248 PLANT NOTES FROM TEMISCOUATA COUNTY

LILIACE^

Streptopus amplexicaulis (L.) DC. Notre Dame du Lac.

Streptopus roseus Michx. Same locality.

Smilacina trifolia (L.) Desf. Same locality.

Maianthemum Canadense Desf. Same locality.

Tofieldia glutinosa Willd. A form with pedicels four to six lines long

in fruit. Grand Falls.
Clintonia borealis (Ker.) Raf. Notre Dame du Lac.

JUNCACE^

J uncus effusus L. Notre Dame du Lac.

Juncusfiliformislu. Same locality.

/. Canadensis Gay, var. coarctatus Engelm. Same locality.

TYPHACEiE

Sparganium simplex Huds., var. Nuttallii Engelm. ( ?) Same locality.

ALISMACE^

Sagittaria variabilis Engelm., var. hastata Engelm. Notre Dame
du Lac.

NAIADACE^

Potamogeton natans L. Same locality.

Potamogeton Pennsylvanicus Cham. (P. Claytonii Tuckerm.), fide
Morong.

ERIOCAULE^

Eriocaulon septangulare With. Dube's Lake, Temiscouata.

CYPERACE^

Carex rostrata With., var. utriculata Bailey. Notre Dame du Lac.
Carex hystricina Muhl. Same locality.
Carex flava L. Same locality.
Carex scoparia Schk. Same locality.

GRAMINE^

Spartina cynosuroides Willd. Notre Dame du Lac.
Deyeuxia Canadensis (Beauv.) Hook. Same locality.
Glyceria Canadensis (Michx.) Trin. Same locality.
Elymus mollis Trin. Cacouna.

PLANT NOTES FROM TEMISCOUATA COUNTY 249

CONIFERS

Thuja occidentalis L. Grand Falls.

EQUISETACEiE

Equisetum sylvaticum L. Notre Dame du Lac.

FILICES

Asplenium Filix-fcemina (Swartz) Bernh. Notre Dame du Lac.
Phegopteris polypodioides Fee. Same locality
Phegopteris Dryopteris (L.) Fee. Same locality.
Aspidium spinulosum Swartz. Same locality.
Cysiopterisbulbifera (Swaitz) Bernh. Grand Falls.
Aspidium marginale Swartz. Same locality.
Osmunda Claytoniana L. Notre Dame du Lac.

OPHIOGLOSSACE^

Botrychium Virginianum Swartz. Notre Dame du Lac.

LYCOPODIACEiE

Lycopodium clavatum L. Notre Dame du Lac.
Lycopodium complanatum L. Same locality.
Lycopodium dendroideum Michx. Same locality.

PLANT NOTES FROM TADOUSAC AND TEMISCOUATA
COUNTY, CANADA^

By John I. and Alice R. Northrop

The little village of Tadousac is beautifully situated at the junction
of the Saguenay and the St. Lawrence. The latter river here forms
quite a deep bay, on the curving shores of which stands the main part
of the village, while on the jutting rocky point that separates the bay
from the mouth of the Saguenay, are the picturesque cottages of the
French fishermen and half-breeds. Just south of the village the banks
become very steep and here, far above the present beach, are two, and
in some places three, very fine terraces, stretching around the point of
the bay and for some distance down the St. Lawrence. Back of all
rise the lofty hills of the Saguenay. Our first walk was out on the
headland at the mouth of the latter river, and here, at the level of the
sea, we found many of the alpine plants of Mt. Washington; among
these were Potentilla tridentata, Euphrasia officinalis, and Vaccinium
Vitis-Idcea. Great mats of Empetrum nigrum grew on the rocky
banks and beautiful Campanulas nodded from the crevices. We were
interested in observing that here Campanula rotundifolia bore only a
single flower at the summit, and resembled but slightly the large and
many-flowered plants, two feet in height, that we had collected in
Crawford Notch, and also very different apparently from the little
blue-bells, three inches high, that grow near the Lake of the Clouds on
Mt. Washington, and bear but a single erect flower, of a very dark blue
color.

We found much the same difference in the forms of Euphrasia
officinalis. We had first collected it a few weeks previous, near Oakes'
Gulf on Mt. Washington, and there, though mature, it was so tiny
that in order to find it, we had literally to get down on our hands and
knees, and go over the ground inch by inch. At Tadousac we collected
the usual small bushy Canadian form, and a week or two later, at Lake

» Bull. Torrey Botanical Club, Vol. XVII, No. 2, February, 1890.

250

PLANT NOTES FROM TADOUSAC 251

Temiscouata, found some specimens about 8 inches high. Besides
this alpine flora, there were many maritime plants. Plantago maritima
and Triglochin maritimum were common, Tissa salina grew on the
tide flats, and Lathyrus maritimus over the rocks on the beach. But
most beautiful of all were the great spreading clusters of Mertensia
maritima, with its graceful trailing branches, glaucous foliage, and
showy blue flowers.

Higher up on the rocks were great patches of Kalmia angustifolia,
making masses of pink bloom; with these grew Ledum latifolium and
Cassandra calyculata, both in fruit, while shading these tangled thickets
were Pinus Banksiana, the white spruce, and the black spruce. Trail-
ing over the rocks we found Juniperus Sabina, var. procumbens, in
fruit, and the prostrate variety alpina of J. communis.

The yellow blossoms and curious inflated pods of Rhinanthus
Crista-galli were common near the shore, and in the wet grassy spots
Microstylis unifolia grew plentifully. Along the Saguenay we collected
Ranunculus Cymbalaria, and in the crevices of the rock Aspidium
fragrans was found growing in large tufts, still holding last year's
leaves, and Corallorhiza innata. Here we also collected Comandra
livida and Goodyera repens.

On the border of a small pond near Tadousac, Polygonum amphi-
bium was found, and in the bogs near the shore we saw Eriophorum
gracile, and with it the purple-flowered Potentilla palustris; also Calla
palustris, but the flowers had already gone, and we could only get the
fruit. Although it was the last week in July, we noticed that the little
birches on the hillsides near Tadousac had begun to turn yellow, show-
ing that the brief Northern summer was already on the wane.

A sail of three hours brought us to Riviere du Loup, and the next
morning we botanized along the banks of the rapid-flowing stream that
gives the town its name. In one place it makes a plunge of about 40
feet over the edges of the dark slaty rock which forms its bed, and here
under the hemlocks we found Moneses uniflora, the bright blue berries
of Clintonia, Coptis trifolia, and Halenia deflexa.

Our train left Riviere du Loup about noon, and we were soon
speeding towards Lake Temiscouata, some fifty miles to the southeast.
At first poplars, birches, and spruces were plenty, but as we neared the
lake, we passed through great swamps of Thuja occidentalis, some of
the trees being the largest we had ever seen. In the cleared places were

252 PLANT NOTES FROM TADOUSAC

large patches of Cnicus arvensis, Epilobium spicatum, and great
masses of the gorgeous yellow heads of Sonchus arvensis, while in
places Vicia Cracca covered the grass and low shrubs with a mantle of
blue. After a ride of nearly three hours, we alighted at our destination,
the little village of Notre Dame du Lac, prettily situated on the curving
shore of Lake Temiscouata.

Two lists of plants from this region have already appeared in the
Bulletin, the first by John I. Northrop, in November, 1887, the
second by Henry M. Ami, in May, 1888.

On my previous visit, in one place near the shore, I collected an
exotic labiate, which was not determined as the specimen was in fruit,
but this year we found the same species, now evidently well established
and holding its own in a dense growth of Galeopsis Tetrahit and A ma-
rantus. It has curious one-sided spikes of blue flowers, and a peculiar
lemon-like odor. We sent some specimens to Dr. Britton, who wrote
in reply that it was Elsholtzia cristata, a native of Siberia, and belonged
near our genus CoUinsonia. We found it growing in a clearing near
a mill, and there is no doubt but that it has been introduced. Still,
there are no records of its ever being cultivated, and how it reached
this out-of-the-way spot is a mystery.

Near by on the banks of a brook Lister a convallarioides grew
abundantly, and on the hillside above we found a few specimens of
Pyrola chlorantha. Great beds of Cornus Canadensis grew here, as
almost everywhere in Canada, and we noted that here, as elsewhere, we
could not find a single four-leaved plant that bore a flower. We had
been struck with this fact while botanizing in the White Mountains,
and had made it a special point of observation all summer, but out of
the hundreds of plants which we examined not one exception did we
find to this rule.

One day we took our canoe and paddled some three miles down the
opposite shore of the lake and here, growing among the moss along a
little brook, we found the rare Selaginella spinosa, looking very much
like a delicate Lycopodium. Farther up the lake on the way to Mt.
Wissick we found a number of plants of Equisetum littorale. The speci-
mens have since been carefully examined and compared with those so
named by Dr. Morong, and they are undoubtedly the same species.
On our way to Mt. Wissick we passed over a shallow portion of the
lake, and a better place to collect water plants could not readily be

PLANT NOTES FROM TADOUSAC 253

found. The bottom was in some places gravelly and in others of sandy
day. The water was beautifully clear, but two or three feet in depth.
We found here five Potamogetons : P. pectinatus, P. gramineus, P.
amplifolius, P. natans, and P. perfoliatus, the delicate Ranunculus
aquatilis, var. trichophyllos, and two species of Chara. Here and there
the spikes of the great bulrush, Scirpus lacustris, rose six or seven feet
above the surface of the water.

Most beautiful of all were the Myriophyllums; one of these, ver-
ticillatum, grew near our camping place in great masses, and we could
look down into the clear water and trace the long, feathery, unbranched
stems until they reached the sand, some seven or eight feet below.

Another species, more slender and growing in much shallower
water, we have since determined as Myriophyllum alterniflorum.
Attention has been called to this interesting species by Dr. Morong in
a previous number of the Bulletin (xiv. 51), where he states that it
has been found in Lake Memphremagog, Canada, by Mr. J. R.
Churchill, but previously had been collected no nearer to us than
Greenland.

Well pleased with our morning's spoils, we paddled on a mile or
so farther, until we found a good camping place, when we dis-
embarked and soon started for the top of la grosse montagne.

Our trail led up through the usual growth of balsam, white spruce,
and Betula papyrifera; here and there the ledges were luxuriously
carpeted with Cladonia rangeriferina. Higher up we found dozens of
specimens of Habenaria Hookeriana, and even more abundant was
Corallorhiza multiflora. We also found some few spikes of C. innata.
About 550 feet above the lake we found some plants of EpigcEa re pens
(the first we had seen in Canada) surrounded by Vaccinium Pennsyl-
vanicum, Kalmia angustifolia, 2ind Ledum latifolium, and shaded by
black and white spruces. Among the crannies of the upper ledges we
found Woodsia Ilvensis, W. hyperborea, PellcEa gracilis, and Aspidium
fragrans.

The summit was shaded by very fine specimens of the red pine,
Pinus resinosa. The next day we worked up our plants, collected
some fossils from the ledges at the foot of the mountain, and later on
in the afternoon started back for Notre Dame du Lac. The few
remaining days we spent there were devoted to grasses and sedges —
there was not a very great variety of either, however, and their names
will be found in the appended list.

254

PLANT NOTES FROM TADOUSAC

The following list contains the plants collected that have not been
mentioned in the former lists. Unless otherwise stated the locality
is in the vicinity of Notre Dame du Lac.

Ranunculus re plans.
Coptis trifolia.
Nymphoea advena.
Nasturtium palustre.
Silene noctijiora.
Hypericum ellipticum.
Trifolium hybridum.
Astragalus orohoides,

var. Americanus.
Amelanchier Canadensis,

var. rotundifolia.
Chrysosplenium Americanum.
Ribes prostratum.
Myriophyllum verticillatum.
Myriophyllum alterniflorum.
Osmorrhiza brevistylis.
Heracleum lanatum.
Carum Carui.
Galium asprellum.
Erigeron Philadelphicum.
Gnaphalium uliginosum.
Lobelia Dortmanna.
Epigcea repens, Mt. Wissick.
Pyrola elliptica.
Pyrola rotundifolia.
Pyrola chlorantha.
Hypopitys Monotropa.
Fraxinus sambucifolia.
Veronica serpyllifolia.
Veronica scutellata.
Verbascum Thapsus.
Mentha sativa.
Elscholtzia cristata.
Apocynum androscemifolium.
Chenopodium capitatum.
Fagus ferruginea.
Polypodium vulgare.
Pellcea gracilis.
Aspidium fragrans.
Woodsia Ilvensis, Mt. Wissick.
Woodsia hyperborea, Mt. Wissick.

Listera convallariodes.

Microstylis unifolia.

Corallorhiza innata.

Goodyera repens.

Habenaria orbiculata.

Trillium erectum.

J uncus nodosus.

Juncus tenuis.

J uncus bufonius.

Potamogeton amplifolius.

Potamogeion gramineus.

Potamogeton perfolidtus.

Potamogeton pectinatus.

Car ex intumescens.

Car ex retrorsa.

Car ex stipata.

Carex tentaculata.

Scirpus atrovirens.

Eriophorum cyperinum.

Dulichium spathaceum.

Eleocharis palustris.

Eleocharis ovata.

Phleum pratense.

Agrostis alba.

Agrostis hiemalis.

Deschampsia ccespitosa.

Poa serotina.

Bromus ciliatus.

Agropyrum repens.

Taxus baccata, var. Canadensis.

Abies Mariana.

Equisetum arvense.

Equisetum limosum.

Equisetum littorale.

Equisetum pratense.

Equisetum scirpoides.

Botrychium Virginicum, var. gracile,

Mt. Wissick.
Ly CO podium annotinum.
L. lucidulum.
Selaginella spinosa.

NOTES ON THE PLANT DISTRIBUTION OF MT.
WASHINGTON, N.H.

By John I. and Alice R. Northrop

In the course of a week's botanizing on the summit of Mt. Washing-
ton in the early summer of 1889 (July 2-9), a fairly complete collec-
tion of its flora was made, largely with the view of ascertaining the
distribution of the plants in relation to altitude. The respective
elevations were determined by means of an aneroid barometer. The
following pages give a brief resume of the results obtained.

The base of the mountain, where the cog railway begins, was made
out to be approximately 2500 feet. Here the flora was about the
same as in the valley. During the first five hundred feet of elevation the
mountain ash {Pyrus sitchensis (Roem.) Piper), Acer spicatum Lam.,
Acer pennsylvanicum L., Prunus pennsylvanica L. f,, Picea mariana
(Mill.) B. S. P., and Ahies balsamea (L.) Mill, were noted. Of these
the striped maple {Acer pennsylvanicum) was not seen above 3000 feet
of elevation, Betula lutea Mich, f., and Acer spicatum were infrequent
in the second five hundred feet and did not reach the 3500 limit, while
Prunus pennsylvanicawsiS abundant up to 3500, but was entirely want-
ing above 4000 feet altitude. Pyrus sitchensis, at first very plentiful,
was much less common at 4000 feet, and at 4300 feet was noted only
as a straggling bush. A single specimen was found at an altitude of
5700 feet near the trail leading into Tuckerman's Ravine. Betula
alba L., var. papyrifera, was abundant from 3500 to 4000 feet, but was
much less frequently met with beyond the 4300 limit, and higher up was
replaced by the variety minor (Tuckerm.) Fernald. This, as a low
straggling bush, held its own far above the timber line, while about
the summit it in turn gave way to Betula glandulosa Michx.

The two conifers, Picea mariana and Abies balsamea, were con-
spicuous throughout, and from 4300 feet upward were almost the only
woody plants to be found. They grew as stunted shrubs, two or
three feet in height, at an altitude of 5000 feet, and even above that a
s 255

256 PLANT DISTRIBUTION OF MT. WASHINGTON

few storm-beaten specimens, hardly a foot in height, were occasionally
seen. One plant, about fifteen inches in height, was noted in a shel-
tered spot near the Crawford bridle-path, at an altitude of 6000 feet.
Near the Lake of the Clouds, at about 5000 feet, clumps of the black
spruce were found with stems three or four inches in diameter, but
with the branches pressed flat against the ground. As a rule, however,
the balsam was the first to be noted as we descended from the summit.

Various trips were made to the Great gulf or Gulf of Mexico,
Oakes' gulf, Tuckerman's Ravine, the Lake of the Clouds, Bigelow's
lawn, and the alpine garden. The plant altitudes were carefully
noted on each descent, and the succession of flowers was almost
invariably the same. On the very summit, 6300 feet, a thrifty speci-
men of that hardy traveler, the dandelion, was in full bloom and
apparently as much at home as on the roadsides several thousand
feet below. Among the rightful owners of the rocky soil the follow-
ing always headed the lists: the Greenland sandwort {Arenaria
groenlandica (Retz) Spreng), the mountain cranberry {Vaccinium
Vitis-Idcea L., var. minus Lodd), and Potentilla tridentata Ait. Grow-
ing with these on some parts of the summit were the alpine goldenrod
(Solidago Cutleri Fernald) and Diapensia lapponica L. The con-
vex clumps of the latter were plentiful in some places, but it was al-
ready in fruit, and only after much searching were one or two belated
blossoms found. Within a radius of a few hundred feet from the
summit were noted the yellow mountain avens {Geum Peckii Pursh.),
common in the grassy places, the two bilberries (Vaccinium ccBSpilosum
Michx. and Vaccinium uliginosum L.), Silene acaulis L., the alpine
azalea (Loiseleuria procumhens (L.) Desv.), and the Lapland rose bay
{Rhododendron lapponicum (L.) Wahlenb.), the latter in ^fruit only.
Although it was late for them, we were fortunate in discovering two
or three patches of the mountain heath {Phyllodoce ccerulea (L.) Bab.),
still covered with the purple bells and clinging to the ledges. Cassiope
hypnoides (L.) Don. was also found in bloom in a few places. On
or very near the summit we also collected the sedges, Carex capitata
L., Carex rigida Good, var. Bigelowii (Torr.) Tuck., Carex capillaris
L., and the wood-rush (Luzula spicata (L.) DC).

All the plants above mentioned were found growing in dry places
among the rocks of the cone, while the following were collected in or
along the brooklets which run down the sides of the Great gulf,

PLANT DISTRIBUTION OF MT. WASHINGTON 257

Tuckerman's Ravine, and Oakes gulf. The vegetation was most luxu-
riant about the streamlets that trickle down the slope at the head of
Tuckerman's Ravine. Here at an elevation of 5300 or 5400 feet
the bright green plaited leaves of Veratrum viride Ait. and the tall
stems and large umbels of the cow-parsnip (Heracleum lanatum
Michx.) first caught the eye. The two species of twisted-stalk
(Streptopus amplexifolius (L.) DC. and S. roseus Michx.) were both
abundant ; while Castillej a pallida L., var. septentrionalis (Lindl.) Gray,
raised its creamy blossoms everywhere. A little farther down the
fragrant spikes of Habenaria dilitata (Pursh.) Gray were found in the
shade of the Veratrum. Spircea salicifolia L., the meadow sweet, was
also noted here. Of the smaller plants the yellow rays of the arnica
{Arnica mollis Hook., var. petiolaris Fernald) and the bright blue flow-
ers of Veronica alpina L.,var. unalaschensis C. and S., were conspicuous.
In moist grassy places everywhere on the mountain Houstonia ccerulea
grew luxuriantly, although it belied its name by bearing pure white
flowers without a trace of blue. (It is now known as variety Faxanorum
Rose and Moore.) The sleeping position of this plant was interesting.
The flowers were partly closed and bent down on their slender stems,
all turning in the same direction, to the northeast in all the plants
we examined. The Northern chickweed {Slellaria borealis Bigel),
Epilobium Hornemanni Reichenb., Polygonum viviparum L., the
mountain sorrel {Oxyria digyna (L.) Hill), and the little alpine violet
(Viola palustris L.) were collected in or along the brooks, as were also
the less common alpine cress (Cardamine bellidifolia L., var. laxa
Lange) and the rare Saxifraga rivularis L.

Of the plants usually found at a much lower altitude we noted the
gold-thread (Coptis trifolia (L.) Salisb.) still in bloom, Ranunculus
abortivus L., Rubus triflorus Richards, Trientalis americana (Pers.)
Pursh., Ribes lacustre (Pers.) Poor, Ledum grcenlandicum Oeder., and
Clintonia borealis (Ait.) Raf. In all our descents Aspidium spinu-
losum (O. P. Miiller) Sw. was the first fern to make its appearance
at an altitude of over 5000 feet, Phegopteris polypodioides Fee. the
second, and Asplenium Filix-foemina (L.) Bernh. the third.

The conditions seemed very similar on all parts of the summit, yet
many of the alpine and subalpine species were very locally distributed.
For example, among the willows, Salix herbacea L. was exceedingly
common in the Great gulf, but was only represented elsewhere by a

258 PLANT DISTRIBUTION OF MT. WASHINGTON

very few specimens in Oakes gulf and the alpine garden. On the
other hand, Salix argyrocarpa Anders., was common in Tuckerman's
Ravine and elsewhere, but none were found in the Great gulf, while
Salix Uva-ursi Pursh. was plentiful throughout. In Tuckerman's
Ravine Salix argyrocarpa and Salix phylicifolia L. grew side by side
and with them a third, a hybrid between the other two. This had
been discovered by Mr. Edwin Faxon, the well-known botanist and
collector, whom we had the pleasure of meeting on his annual visit
to the mountain.

Further instances of local distribution were that of Saxifraga
rivularis which we believe has never been collected elsewhere than the
one little brook in which we found it, and Gnaphalium supinum L.,
which is still more rare, growing only, so far as we were able to dis-
cover, in two small stations, each but a few feet in extent. We were
fortunate in finding specimens of this alpine cudweed, although they
were only a little over an inch in height.

In company with Mr. Faxon, we found in the alpine garden
after some search a small patch of the alpine bearberry {Arctosta-
phylos alpina (L.) Spreng.), where he remembered having seen it
some years before. Silene acaulis was found in but one locality near
the summit, although quite plentiful there, and Potentilla Rohhinsiana
Oakes was noted only near the Lake of the Clouds. The little
eyebright {Euphrasia Oakesii Wettst.) was met with in only one
locality of small extent which was pointed out to us by Mr. Faxon.
Again Prenanthes nana (Bigel) Torr. seemed to be entirely confined
to one side of the mountain, and Prenanthes Bootii (DC.) Gray to
the other. The cloudberry {Rubus Chamcemorus L.) was collected
on the summit of neighboring Mt. Clinton, but not on Mt. Washing-
ton itself. There seems to be no reason why the distribution of
these plants should be so restricted, as in many instances the
conditions are apparently favorable to their spread, but perhaps
this remnant of a more Northern flora finds it sufficiently difficult to
hold its own. During the week spent on the mountain sixty species of
plants in flower were collected above an altitude of 4600 or 4700
feet. These represented twenty-three orders and forty-six genera,
to which are to be added twenty additional species, some of which
were in fruit, others in leaf only, making a total of eighty species
growing above the altitude mentioned.

A STUDY OF THE HISTOLOGY OF THE STEM OF THE
WAX PLANT, HOYA CARNOSA (L.) R. Br.

By John I. Northrop

In the fall of the year 1886 Dr. N. L. Britton suggested that I
study the anatomy of the wax plant, Hoya carnosa, as it appeared
to be hitherto undescribed and possessed interesting peculiarities of
structure.

From that time to the present I have had the investigation in
hand and the following pages present the result.

History

I have been unable to find any account of the discovery of the
wax plant or mention of the name of the person who first collected it.
Linnaeus named it Asclepias carnosa; and Loureiro, under the name
of Stapelia Chinensis, describes a plant which he thinks may be the
same as that called Asclepias carnosa by Linnaeus.

In 1809 Robert Brown divided the order Apocinece, retaining the
old name for one part and calling the other the Asclepiadece. In the
latter order he named for Thomas Hoy, an '* intelligent and success-
ful cultivator," for some time gardener to the Duke of Northumber-
land, the genus Hoya, in which he placed the Asclepias carnosa of
Linnaeus and the Stapelia Chinensis of Loureiro, uniting them under
the name the plant has since borne, Hoya carnosa (L.), R. Br.

Brown considered the wax plant the type of the genus, and states
that it "probably includes several species which can only be determined
from living specimens." As its habitat he gives tropical Asia and
New Holland.

De Candolle, in his " Prodromus," gives for localities "Eastern
India, tropical Asia, around Canton" (quoting Loureiro), and ques-
tions New Holland. George Bentham, in the "Flora Hongkongensis,"
for localities gives "Mount Victoria, Champion; also in adjoining
continent, not known for certain out of South China, but probably

259

260 HISTOLOGY OF STEM OF THE WAX PLANT

extending to Sikkam, as one specimen was found, without any indica-
tion of its precise station, among Dr. Hooker's collection from that
country."

The plant was described by Schrank as Schollia carnosa, and by
Jacquin fils as S. crassifolia; these two names are probably both
subsequent to that given by Brown. Miller, in the " Gardener's Dic-
tionary," calls the plant the *' Fleshy-leaved Swallow Wort." Loudon,
in the ^'Hortus Britannicus," states that it was introduced into Britain
in 1802 ; and in the '* Encyclopedia of Plants " that " one or two plants
placed when in flower in a vinery of ripe grapes will entice the wasps
from eating the ripe fruit." In the Botanischer Jahresberichf, 1881,
it is stated that the honey of Hoya kills bees.

Darwin, in his investigations on climbing plants, speaking of the
gradation of twining plants to simple root climbers, mentions that
Bignonia Tweedyana and Hoya carnosa revolve and twine, but like-
wise emit rootlets which adhere to any fitting surface. The plants
I have had showed no aerial roots, with the exception of a large one
that had been trained over a wooden support, and developed a number
of rootlets where the frame and the plant had been for a long time in
contact.

General Description of the Stem

• PL I, Fig. I. PL II, Fig. 4

For convenience in description, the stem of Hoya carnosa may be
divided into the following zones of tissues, which are best seen in a
section about 4 mm. in diameter. Proceeding from the periphery
to the center we have : —

1. The epidermis with its hairs, stomata, etc., composed of one
row of polygonal cells with the thick cuticle on the outside. The
lateral walls are much thinner, and the cell is completed by one or two
walls which are contiguous with the next row of cork cells.

2. The cork, composed of one or more rows of rectangular cells.

3. A broad band of parenchyma containing many intercellular
spaces.

4. A ring of sclerotic cells, called by Davids the sclerotic girdle,
in a stem of the above size, generally consisting of two rows of cells,
which are lignified, and resemble closely the sclerotic cells of the pear.

5. A broad band of parenchyma, similar to the third ring,

HISTOLOGY OF STEM OF THE WAX PLANT 261

containing the bast fibers which are scattered in bundles just within
the sclerotic girdle.

6. The cambium composed of thin-walled cells.

7. The wood containing the medullary rays, ducts, etc.

8. The pith made up of parenchyma with intercellular spaces.

9. A few sclerotic cells in the center of the pith similar to those
of the girdle, but in a section of a young stem, slightly larger, being
circular in outline; while those of the girdle are compressed, having
their shorter axis on the radial line.

A section corresponding to the above description may serve as a
type with which to compare an older and a younger stem. In the
younger portion of the stem (the first int^rnode) the most noticeable
features are the absence of the sclerotic girdle and cork cells, and
the increased number of hairs and stomata.

In the older portions of the stem the principal changes consist in
the relative space occupied by the different tissues and their modifica-
tions. For example, in a section of a stem 10 mm. in diameter we find
the epidermis missing, and the cork, which is popularly known as
" bark," beginning to part in places. The space between the sclerotic
girdle and the cork has grown much narrower, and is dotted here and
there with many sclerotic cells like those of the pith. The sclerotic
girdle is composed of cells irregular in shape, and what we may call
their tangential diameter is now much greater than their diameter
measured on the radius of the stem. The ring of parenchyma con-
taining the bast has grown narrower and now contains sclerotic cells
here and there.

The wood has increased and occupies a proportionately much
greater part of the stem. It is very irregular in outline and is often
penetrated by wedge-shaped masses of parenchyma. The ducts are
prominent owing to their increased number. The pith is contracted,
and in its center is a mass of sclerotic cells. Just within the ring of
wood will be noticed collections of small cells forming well-marked
patches around the pith. These I have been unable to find men-
tioned in any work at my command, unless they are what Davids,
in his "Ueber die Milchzellen der Euphorbiaceen, etc.," calls soft
bast bundles (" Weichhast hundeV^). Crystals and milk cells will be
considered in treating the different rings of tissues.

262 HISTOLOGY OF STEM OF THE WAX PLANT

Epidermis

PL I

If the stem of a young plant is examined, it will be found covered
with a growth of velvety hairs, and numerous adventitious buds will
be seen between the internodes. Portions of the epidermis may be
easily obtained by drying a piece of stem that has been in alcohol,
or they may readily be pulled off from an older part of the plant. It
is composed of cells varying from 15 /it to 100 fx in length, and from
10 to 30 ft in width, with their greatest diameter parallel to the axis
of the stem. The depth of the epidermal cells is more constant than
their width or length, being about 20 to 30 ^l. Their outer wall is
about 5 A'' in thickness. Treatment with concentrated sulphuric acid
dissolves the rest of the cell and leaves the cuticle as shown on pi. I,
fig. 5. Caustic potash causes the outer wall to swell slightly, when the
striated cuticularized portion can be seen covered by the more re-
fractive cuticle proper. Schultze's solution stains the epidermis orange-
rufous, the stain showing over the same portion of the cell wall that is
left after the action of sulphuric acid. Hence we may conclude that
the external and part of the lateral walls of the epidermal cells are
cutinized.

Trichomes

PL I, Fig. 2

The Trichomes are simple hairs formed by the prolongation of one
epidermal cell, and when young are filled with protoplasm. Concen-
trated sulphuric acid does not dissolve them, and Schultze's solution
stains them the same color as the outer wall of the epidermal cells-
Hence I conclude that their cell walls are cuticularized. Their aver-
age length is about .4 mm. They are divided by septae into three or
more cells. At the base of the hair the wall is equal in thickness to
the outer one of the epidermal cells, but soon becomes much thinner,
and the greater portion of the hair is bounded by a wall about 2|- /^ in
thickness.

Stomata
PL II, Fig. 3^

These are most numerous on the youngest portions of the ^^<^em,
and, with one exception, all I have seen have their opening par^ ^1

HISTOLOGY OF STEM OF THE WAX PLANT 263

to the axis of the stem. It is difficult to estimate the number in a given
space, as they are very irregularly distributed and the proportion
varies with the age of the stem. In a square millimeter of epidermis I
have counted ten stomata in one case and in others but one; perhaps
five would be a fair average for the younger portions of the stem.

Papilla

These are developed on the older portions of the stem and appear
to be formed by a thickening of the central portion of the outer wall
of the cells of the epidermis.

Cork

As stated before, the outer wall of the epidermis is deeply cutinized,
but in the younger portions of the stem we find no cork. As the stem
increases in age and the lateral walls of the epidermal cells become
suberized, the next cell, the cork cambium or phellogen, divides
tangentially, the outer half forming the first cork cell which is suber-
ized in the same manner as the epidermis. The inner half of the cell
again divides, and in this way cork is continually formed. The
development of the cork is shown on pi. II, fig. 4, in which p is
the phellogen which divides to form the cork cells c. The alteration
that takes place in the cell walls changing them from cellulose to cork
proceeds from the outer wall to the inner one.

"It is not possible to account satisfactorily for these substances,
subterin and lignin, in cell walls. They are probably the result of
modification of the cell wall and not of interfiltration." ^ This state-
ment is supported by the manner in which the phellogen cells of Hoya
carnosa become suberized. Treatment with sulphuric acid shows
this very plainly. In the younger portions of the stem the epidermis
alone is left after the parenchyma is destroyed, then the lateral and
finally the inner wall remains, the suberization always taking place
toward the phellogen. Now it has been demonstrated that in coni-
fers the cells become suberized after they have lost their protoplasmic
contents, and this would seem to be the case in the cork of Hoya
carnosa.

As soon as the cork is formed, the cells on the outside are shut off
from further supply of nourishment, as far as we know, and if the

^ " Physiology of Plants," S. H. Vines, p. i8.

264 HISTOLOGY OF STEM OF THE WAX PLANT

suberization was the result of interfiltration, it would seem as if it
should proceed /row the growing portion of the plant, and not toward
it from those that are dead. If the suberization was caused by the cell
walls absorbing the contents of the cell, it is but fair to suppose that
they would become suberized equally, although the opposite might
be the case. However it may be, in Hoya carnosa it would seem as
if some change which we call suberization took place in the cell wall,
caused by being cut off from the supply of nourishment.

The Outer Band of Parenchyma

This is most prominent in the younger parts of the stem and is grad-
ually reduced in thickness as the plant increases in age, till in the older
portions it is equaled in depth by the cork. It is composed of spheri-
cal cells having thin cellulose walls, varying in diameter from lo to
50 /x and increasing in size from the cork to the sclerotic girdle. The
cell walls dissolve readily in sulphuric acid and are turned blue by the
action of Schultze's solution. In the older portions of the stem they
become irregularly thickened and then present a pitted appearance.

This band contains most of the chlorophyll and starch, also
protoplasm, sclerotic cells, milk tissues, crystals of oxalate of lime,
and the coloring matter of the younger portions of the stem. The
last is a rose-purple, apparently homogeneous fluid. Alcohol dis-
solves it rather slowly but much faster than it does the chlorophyll.
The addition of caustic potash changes the color to a sea-green, which
is quite distinct from the yellowish green of the chlorophyll. The
contents of all the cells of the bark are changed in color to an apple-
green, which decreases in intensity toward the cambium and deepens
in the vicinity of the cells that before contained the purple coloring
matter. Hence it would appear that this material is contained in all
the cells of the bark, but only in the outer ones does it become visible.
The green color produced by potash gradually fades and leaves the
chlorophyll granules apparently unaltered in color and form. Con-
centrated hydrochloric acid causes the color to change to a rich wine-
purple, which is confined to the cells containing the coloring matter
and does not spread through the other cells like the green caused by
potash. The color lasts perhaps ten minutes and then fades. Strong
sulphuric acid has the same effect, but the walls of the parenchyma
swell and are gradually destroyed.

HISTOLOGY OF STEM OF THE WAX PLANT 265

Schultze's solution at first changes the color to wine-purple and
then shows the cells that contained it to be quite filled with starch.
I have been unable to identify this substance with any description
at my command of the coloring matter of plants. It is probably
complex in structure, and further investigation is needed by some one
more skilled in micro-chemistry than myself.

Crystals

These are composed of oxalate of lime and are contained in con-
siderable numbers in this band of tissue, mostly outside of the sclerotic
girdle. They are of a general spherical shape, presenting many sharp
points over the surface as shown in pi. II, fig. 6. It has been stated ^
that they are inclosed in a cellulose wall and supported by threads of
cellulose ; this I have been unable to confirm, but think from the way
in which they act when treated with hydrochloric acid or Schultze's
solution that it is so. They certainly have a nucleus of organic
matter.

Milk Cells

PI. I, Fig. 7

Davids, in 1872, published a pamphlet containing the results of
his investigations of the milk cells of the EuphorbiacecB, Moracece,
Apocynacece, and AsclepiadacecB, and there gives references to the litera-
ture on the subject. The Apocynacece and Asclepiadacece he treats
together, taking Nerium oleander as the type of one, and Hoya car-
nosa of the other, stating that in the latter the milk cells are almost
exclusively confined to the parenchyma inside the sclerotic girdle.
He gives their diameter from 18 to 21 ^l, states that they are moder-
ately branched, have no septa, and that their ends are closed. Also
that they are formed in the growing point and are developed simul-
taneously in the hark and in the pith. They are not cell fusions hut cells
which hy stretching and branching reach a considerahle length in the
intercellular space. They consist of fundamental cells and are not to he
considered homologous with the true hast nor as part of the hundle sys-
tems (Gefasshiindel Systeme). They do not anastomose with each other
and therefore huild no " system'' in plants. As far as my investigation
has gone the above observations are correct, but the milk cells are

^ Botanischer Jahresbericht, 1880, vi, p. 31.

266 HISTOLOGY OF STEM OF THE WAX PLANT

found outside of the sclerotic girdle more frequently than the above
statement would seem to imply. They are filled with a greenish
yellow fluid distinctly granular in structure. They can be very readily
seen in a longitudinal radial section and also in a tangential one that
passes through the parenchyma of the bark.

Chlorophyll

The chlorophyll is principally contained in the outer border of the
parenchyma and is composed of granules about 5 At in diameter. I
have been unable to decide whether they contain starch, but think
that they do. The inner band of parenchyma also contains chloro-
phyll on its outer boundary, but it is not found in any other parts of
the stem, to my knowledge.

Starch

In the outer band of parenchyma starch is very abundant and
completely fills some of the cells. The granules vary in size, but 5 ^l
would perhaps be a fair average. They do not show any markings as
far as I have been able to observe, except a slight suggestion of a
granular structure. In this band of tissue starch also occurs in the
inner ring of parenchyma, in the wood, pith, and medullary rays, if
what I consider the latter proves to be correct. The occurrence of
starch varies with the treatment the plant has received, for on making
a' section of a plant that had been for some time without water, the
starch had almost entirely left the parts of stem where it was abundant
before. In regard to the protoplasm I can only say that it is contained
in the cells of this band, but I have not attempted to investigate its
properties, as the object of this paper is to describe the structure of
the plant cells rather than their contents.

Sclerotic Cells

These will be considered when treating the sclerotic girdle. It is
only necessary to state here that they occur in abundance just under
the bark and are irregularly scattered through the band of parenchyma.
They increase in number as the plant grows older, and it is in the old-
est portions of the stem that we find them most numerous.

This plant offers better opportunities for the study of these cells
than any known to me. They are to be found in the petiole and leaf

HISTOLOGY OF STEM OF THE WAX PLANT 267

as well as in the stem, and being so abundant, their development can
easily be traced, and the changes that take place in the contents of
the cells that later become transformed offer an interesting field
for investigation. It is difficult to account for their presence. The
sclerotic girdle may be readily supposed to be for the purpose of
strengthening the stem, but these separate cells cannot be considered
in that light, as they are surrounded by parenchyma and simply add
their weight to the stem without contributing to its support. They
do not connect with each other except where two or more happen to
be in contact, and therefore can hardly be considered as part of the
system of the plant. They may better be regarded as the result of
secretions for which the plant has no further use.

Sclerotic Girdle

The sclerotic girdle makes its first appearance in the third or fourth
internode by the lignification of some of the cells of the parenchyma
surrounding the bast, thus forming a broken ring. The intervening
cells soon become lignified, and we have the girdle formed of a single
row of cells which afterward increases to two or more rows.

Now if a young portion of the stem in which the sclerotic girdle
has not yet formed is treated with iodine, the first row of cells sur-
rounding the bast will be seen to be filled with starch forming a distinct
and striking circle of a dark blue color. It is the row of cells surround-
ing those containing the starch that develop into the sclerotic girdle,
and it seems almost impossible to resist the conclusion that the starch
is in some way connected with the growth of the sclerotic cells.

Lignification of the Sclerotic Girdle

The walls of the cells that form the girdle are at first slightly
thickened on all sides, leaving small pores which later form canals
as shown on pL II, fig. 7. I have not been able to decide whether
these ''pores" are holes through the cells or spaces left unlignified^
but I think the latter, as it seems that if they were holes they would
appear before lignification, which, as far as my knowledge extends, is
not the case. The cell walls then increase in thickness either by the
formation of substance on the inside or interfiltration, between the
pores, which thus become canals. In my opinion the cell walls are
formed by layers added to the inside, as they are distinctly striated.

268 HISTOLOGY OF STEM OF THE WAX PLANT

and as the canals grow longer, their ends approach each other, the
portions of the cell becoming wedge-shaped, as shown on pi. II,
fig. 7, and this would be hard to account for on the theory of interfil-
tration, for then as the cell becomes more and more lignified, the inner
portion would have to contract or the canals become closed. Now
while some of the canals do run together and form one, the majority
of them run from the circumference to the center. At last the cell
becomes almost entirely closed, with the exception of the canals before
mentioned.

An interesting problem arises in connection with the sclerotic
girdle. It is formed in the young stem and incloses a portion of the
soft tissues between itself and the wood. Now as the stem becomes
thicker, the girdle must increase its diameter. The cells surrounding
it keep pace with the enlargement of the stem by the simple act of
division. The sclerotic cells do not divide, to my knowledge, for
they are lignified almost to the center, and any division could be very
easily seen. The problem is, how does it grow ? We know that the
tissues of plants are under considerable pressure, as is shown by simply
cutting a stem in two, when the fluids contained will immediately
exude, and this can be explained only by the supposition that they are
squeezed out by the contraction of the tissues surrounding them.
This tension will explain the flattened form assumed by the sclerotic
girdle in the older parts of the stem, for, as stated above, the compres-
sion takes place on the radial line. The difference in shape between
the cells composing the girdle and the sclerotic ones near it is very
marked, the latter being comparatively round. The increase in
tangential diameter the cells receive would seem hardly sufficient to
allow the stem to expand as it does. The following experiment was
made. A section was chosen in which the sclerotic girdle had partly
formed and in which the cells composing it were almost round. The
diameter of the girdle was measured and found to be 2 mm; the
circumference was then calculated and found to be 6.2832 mm. The
cells composing the girdle were counted and found to be about 209
in one case and 226 in another. The former figures give .03 mm., the
latter .0288 mm., as the diameter of the cells composing the girdle. An
average cell of the girdle was then measured tangentially and found
to be .035 mm. in diameter. As the cells are slightly longer in that
direction than radially, these figures are quite close, and we may adopt

HISTOLOGY OF STEM OF THE WAX PLANT 269

the figure obtained by calculation (.03 mm.) without serious error.
Making the same calculation for a girdle 3 mm. in diameter where the
cells were 220 in number, we find their diameter to be .04 mm., while
their measured diameter is .05 mm., the difference being doubtless
due to the overlapping of the cells.

^1 1 ^. 2 X 3.1416 6.2832

Calculation ^ — - — = ^- = .035 mm.

209 209

Hence we see that the elongation of the cell is sufficient to account
for the increase of the girdle thus far. But when we look at a section
in which the sclerotic ring measures 7 or 8 mm. in diameter, we see
that it contains many small round cells as well as the flattened ones,
and in some places the girdle is separated by a cell of parenchyma,
thus suggesting that the increase in growth is not due to the compres-
sion of the cells alone, but that some of them separate, and the spaces
left vacant are filled by parenchyma that soon becomes lignified. This
must take place very rapidly, as I have never found sclerotic cells in
the process of formation in the girdle of the older portions of the stem.

As stated before, the object of the sclerotic girdle is probably the
support of the young stem, as the plant twines but feebly. When the
stem reaches a diameter of ten or twelve millimeters, the girdle no
longer stands out as a conspicuous component of the section, but is
separated in places and appears to be gradually losing its identity, and
I venture to say, if its object is as I have supposed, that in a still larger
stem the girdle, being no longer needed for support, would be found
disintegrated and mingled with the sclerotic cells of the parenchyma.

The Inner Band of Parenchyma

This is composed of cells similar in structure, shape, and size to
those of the outer band, and also contains besides the bast fibers,
sclerotic cells, chlorophyll, crystals, and starch. It is the outer cells
of this band that are filled with starch before the development of the
sclerotic girdle. The bast fibers lie scattered in bundles mostly near
the outer boundary of the band. They are fusiform in shape, about
20 fJL in diameter, their walls are about 6 /-t in thickness, and are very
faintly marked in the manner shown on pi. II, fig. 8.

It is very difficult to determine their average length. None of
those I have been able to isolate with certainty have measured over

270 HISTOLOGY OF STEM OF THE WAX PLANT

3 J mm. The ends of the fibers are generally prolonged into slender
threads, but in some cases the termination is quite abrupt, and they
expand slightly near the end as shown on pi. II, fig. Sb. Sulphuric
acid causes the fibers to swell, and contract violently in the direction
of their length, and finally dissolves them. Schultze's solution changes
them to a deep azure-blue color ; while iodine shows a slight yellow
tinge. They contain a fluid that stains readily with magenta and
iodine, and does not appear to be affected by sulphuric acid. Scat-
tered here and there in this band of parenchyma will be seen crystals,
sclerotic cells, and milk cells (latex cells). These are in all respects
similar to those found in the outer band, and a detailed description
of them need not be repeated.

Cambium

The cambium cells are composed of exceeding delicate cellulose
walls and by successive division form the wood, medullary rays, etc. ,
on one side, and the tissues that surround them, on the other. The
development of these tissues has not been attempted. The cells
are generally about 15 /tt in cross section and 50 yw- in length.

Wood

Fritz Miiller, in an interesting paper in the Botanische Zeitung,
February, 1866, describes the structure of the wood in climbing
plants about Disterro and gives figures showing the irregular form that
it sometimes assumes. He states that these irregularities are due to
the separation of the woody portions or to "strings of soft tissue
drawn through them," the stem becoming thereby more flexible
than if the mass of wood was regularly formed.

In the young stem of Hoya carnosa the outer boimdary of the wood
is quite regular and passes in to the cambium in the ordinary manner.
But as the stem increases in age, the growth is more and more irregular,
till in a stem about 10 mm. in diameter the appearance of a section is
like that shown on pi. II, fig. i, 4; the wood being generally much
thicker on one side than on the other and penetrated in many places
by wedge-shaped masses of parenchyma. I have not been able to
decide whether the wood separates and allows the soft tissue to pene-
trate it, or whether it simply grows irregularly around them, but I
think the former, as in the young portions of the stem the wood makes

HISTOLOGY OF STEM OF THE WAX PLANT 271

an unbroken circle, while in some of the older portions the parenchyma
has reached almost to the pith and the distance separating them is
much thinner than the band of wood in the young stem.

In a tangential section the masses of parenchyma are shown to
be lenticular in shape with their longest axis parallel to the axis of the
stem, looking very much like medullary rays, which I at first took them
to be, as none of the other cells in such a section show the ordinary
appearance so generally characteristic of these tissues. If, however,
we make a radial section, we can distinguish four well-marked forms
of cells: (i) the wood cells proper; (2) the ducts; (3) the cells
of the parenchyma, if the section has been made through a portion of
the stem containing them; (4) rectangular cells, very regular in
form and length, their ends thus forming well-marked lines across
the section. These, I think, are the medullary rays.

Next to the pith will be seen cells with thickened walls giving them
the appearance of a spiral. These latter are best seen in a section of
a young stem. These various tissues with the exception of the paren-
chyma all exhibit the characteristic reactions of lignified cells, staining
with the various reagents used for the purpose of identifying them, such
as phlorglucin and thymol. As far as my investigation has gone, the
ducts contain only air. The medullary rays and parenchyma are
usually filled with starch ; the wood cells sometimes contain the latter,
but are generally empty. The position of the starch varies with the
season of the year and the circumstances under which the plant has
been kept.

Ducts

In a transverse section of the stem these will be noticed scattered
through the wood. Their diameter is much greater than that of the
wood cells, being from 25 to 50 /-t, while the latter are quite constant
in size, having a cross section of about 15 At. The length of the ducts
varies greatly, some being 100 /^ long and others only 10 /x. They
form large canals through the stems, the ends meeting sometimes
at right angles, but generally overlap, making an angle of about 40°
with the direction of their length. Their walls are about equal in
thickness to the walls of the wood cells, but are regularly pitted as
shown on pi. II, fig. 3.

Sachs, in his "Physiology of Plants," states that Theodore Hartig,

272 HISTOLOGY OF STEM OF THE WAX PLANT

twenty or thirty years ago, showed that the "histological elements of
the wood are not in open communication with each other and that the
bordered pits of the wood cells (and the vessels of foliage trees) are
not actually perforated, but closed by fine membranes." This is
confirmed by Sachs, in the experiment he gives on the same page,
namely, by filtering through the wood of a conifer a fine emul-
sion of cinnabar, the cinnabar not penetrating the wood cells. He
states that this is also true of foliage trees. In some thin sections of
wood I thought I could see the connection between the cells, but
could never satisfy myself that there really was one, as it is very
difficult, and in fact a matter of chance, to get a section showing clearly
the edge of the wall of a duct.

The ducts are surrounded by small wood cells and are generally
in groups, sometimes two being contiguous and presenting the appear-
ance of one large one having a longitudinal partition. At the point
where the ends of two ducts meet, the walls are not pitted around a
circle indicating the point of union, but in the central portion of the
transverse wall is a circular plate that can be removed, as some sec-
tions show a space that was formerly occupied by the disk, and in a
slide of macerated wood cells we find these round perforated portions
of the ducts abundant. On pi. II, fig. 3, is figured the end of a duct,
showing the circular space mentioned, and fig. 3a of the same plate
shows the disk that formerly occupied it.

The ducts are formed in the cambium, and while increasing in
thickness, their walls become pitted. These pits are present in both
the tangential and radial faces and also where the ducts meet end to
end. Some portions of the walls, however, are without pits, but their
relation to the position of the cell I have not determined, as I have
only noticed the fact with certainty in longitudinal sections, and, of
course, it is not possible to say what occupied the place before it had
been cut off.

Wood Cells

These are very irregular in shape, and some of the forms they
assume are shown on pi. II, figs. 5-5^. Their diameter is quite
constant, being about 15 A^. Their length varies, as a glance at the
plate mentioned above will show ; some are 5 mm. in length and others
but .15 mm. They may easily be obtained for examination by macer-
ating a portion of the wood, or by boiling a piece in Schultze's solution.

HISTOLOGY OF STEM OF THE WAX PLANT 273

They have rather thick lignified walls that are marked by pits having
a slit-like appearance, the longer axis of the slit forming an angle
with the sides of the cells.

Medullary Rays

The cells of the medullary rays in the plant we are considering
present an exception to those usually found, in that they stand verti-
cally, or with their greatest length parallel to the axis of the stem.
Generally these tissues are well marked in a tangential section by their
lenticular form and the difference in the thickness and composition of
the walls of their cells as compared with those of the other components
of the wood; they show equally well in a radial section the fact that
their greatest length is along the radial line, thus crossing the axis of
the other cells at right angles.

In Hoya, as stated before, tangential sections of the old parts of
the stem show lenticular masses of parenchyma that might be easily
mistaken for medullary rays. The young stem, however, does not
contain them, and if we admit they are the rays, we must concede that
none exist in the younger portions of the plant. If a radial section of
a stem is examined, cells will be noticed that are quite regular in shape,
as shown on pi. I, fig. 6, being prismatic in form, and, as shown in
the section, rectangular in outline and usually quite filled with starch.
Their sides are pitted, as are also the ends. They are generally about
I mm. in length, though sometimes longer or shorter, and about
.015 mm. in diameter. In a cross-section of the stem they may be
easily recognized by the starch they contain and by the pitted plate or
end of the cells, some of which are generally shown. '' But few minute
investigations on the structure of the cells of medullary rays exist, and
many details of structure are still to be discovered." ^

If I am correct in supposing the tissues described above as medul-
lary rays, they resemble the wood cells very closely in all but their
shape. They run from pith to bark and are marked with pits similar
to those of the wood cells. The inner boundary of the wood is formed
by a few long cells with the walls thickened in a spiral or annular
manner. In cutting a section the cells separate very readily and, as it
were, uncoil. They are lignified like the wood, but in cross-section
are circular with the walls slightly thinner.

^ De Bary, "Comparative Anatomy of Ferns and Phanerogams," p. 486.

HISTOLOGY OF STEM OF THE WAX PLANT 275

Pith

The pith is composed of spherical cells from lo to 50 ix in diameter,
being generally filled with starch and having thin cellulose walls
without markings, as far as I have been able to observe. In the center
are a number of sclerotic cells similar to those found in other parts
of the plant, but more regular in shape. These form the axis of the
plant, growing upward in advance of the sclerotic girdle. They are
of the same size as the cells of the pith and are developed like those
of the girdle by the thickening of the cell wall. Crystals also occur
in abundance, and are sometimes surrounded by the sclerotic cells.
They have a nucleus of organic matter and often completely fill the
cell that holds them. Around the pith will be noticed clusters of
small cells presenting the appearance shown on pi. II, figs, i, 4.
These, I think, are what Davids calls Weichhasthundels. They
are about 10 jjl in diameter and 50 ^i in length. These cells are filled
with a colorless fluid and bounded by thin cellulose walls. They
are sometimes arranged around the pith, forming four well-marked
patches, as shown in the figure referred to.

Such is a general account of the structure of Hoya carnosa. If we
glance for a moment at its methods of growth, we at first see the young
stem with the column of sclerotic cells in the center, which, as it
advances, is followed by the girdle which envelops the tissues within
it and furnishes a rigid support to the plant with the least expense of
material, and, if my supposition prove correct, disappearing when the
wood has reached a sufficient thickness to render its presence super-
fluous.

I wish in conclusion to acknowledge my obligations to Dr. N. L.
Britton and others who have kindly assisted me with advice and sug-
gestions.

EXPLANATION OF PLATES

Plate I

Fig. I. Cross-section of young stem showing the different zones of tissues men-
tioned on page 4.

Fig. 2. Hair from young stem. Page 7.

Fig. 3a. Transverse section of a stoma. Fig. 4. Diagrammatic showing develop-
ment of cork, p = phellogen. c = cork. Page 8.

Fig. 5. Epidermis of stem as left after action of sulphuric acid.

276 HISTOLOGY OF STEM OF THE WAX PLANT

Fig. 6. Medullary ray. Longitudinal radi?il section. Page 30.
Fig. 7. Milk cell from outer band of parenchyma.
All figures magnified 200 times linear.

"~~ Plate II

Figs. I and 4. Cross-section of stem showing irregular growth of wood, i X 10

times linear; 4 X 40 times linear, p. 24.
Fig. 2. Tangential section of old stem showing lenticular form of penetrating

parenchyma. X 10 times linear.
Fig. 3. Short duct from old stem showing manner of pitting. 3a, disk that

occupies space above it. X 400 linear, p. 27.
Figs. 5-5^. Different forms of wood cells. X 200 linear, p. 29.
Fig. 6. Crystal from pith of young stem. X 200 linear, p. 11.
Fig. 7. Sclerotic cells from girdle of stem. X 200 linear.
Figs. 8-86. Bast fibers showing manner of striation. X 200 linear.

Plate II.

fc^ « V . >V< oi^Xivcdie Ajc^Xc ck^ t-voJ>-

INDEX TO BAHAMAN PAPERS

The genera of Actinia, Birds, and Crustaceans are given in the following index, and \}s\t. families oi
plants, together with all the new species.

Acanthacaea, 183.

Acanthonyx, 79.

Acasta, 96.

Achelous, 76, 77.

Actsea, 75.

Actinia, 103-117.

Actites, 61.

^gialitis, 61.

Agelaius, 56.

Aiptasia, 103, 117.

Aizoaceae, 153.

Aletris bracteata n. sp., 144, Plate i.

Alpheus, 89-93.

Alpheus hippothoe bahamensis n, var.,

89, Fig.^
Alpheus nigro-spinatus n. sp., 91, Fig.
Amarantaceae, 152.
Amaryllidaceae, 145.
Anacardiaceas, 164.
Anastraphia northropiana n. sp., 190.
Andros, general account, 13, 23, 24, 28, 29,

50, 122.

Previous collectors, 50, 51, 120, 126, 127,

Anguria keithii n. sp., 187, Plate 18.

Anonaceae, 153.

Anous, 64.

Antrostomus, 58.

Apocynaceae, 175.

Ardea, 62, 63.

Arecaceae, 141.

Aristolochiaceae, 150.

Asclepiadaceae, 175.

Athanas ortmanni n. sp., 93, 94, Fig.

Bahaman collection, general summary,

25, 34.
Bahama Islands, general description, 27, 28.

Formation, 46, 47.

Bahaman trip, general account, 1-24.
Balanoglossus, 5.
Batideae, 152.
Bignoniaceae, 182.
Birds, general notes, 48.'

Annotated hst, 50-65.

Bixaceae, 168.
Blacicus, 57.
Boraginaceae, 178.
Bromeliaceae, 144.
Burseraceae, 162.

Cactaceae, 171.

Caesalpinaceae, 156.

CalUchelidon, 55.

CalHnectes, 76.

Canellaceae, 169.

Capparidaceae, 154.

Cardisoma, 72.

Caricaceae, 170.

Cassia Caribaea n. sp., 156, Plate 6.

Catesbaea fasciculata n. sp., 184, Plate 16.

Cathartes, 61.

Caulerpaceae, 129.

Celastraceae, 164.

Ceramiaceae, 131.

Ceryle, 60.

Chastangiaceae, 131.

Characeae, 132.

Chara depauperata n. sp., 132.

Charadrius, 61.

Chenopodiaceae, 75.

Chlorodius, 75.

Chordeiles, 58.

Cladophoraceae, 129.

Coccyzus, 60.

Codiaceae, 129.

Coenobita, 81.

Coereba, 55.

Columba, 61.

Columbigallinas, 61.

Combretaceae, 171.

Commelinaceae, 144.

Compositae, 188-190.

Compsothlypis, 53.

Condylactis, 103.

Coniferae, 134.

Convolvulaceae, 176, 177.

Corallinaceae, 131.

Crassulaceae, 154.

Crossopetalum coriaceum n. sp., 165,

Plate 9.
Crotophaga, 59.
Crudferae, 154.
Crustacea, report on, 69-98.
Cucurbitaceae, 187.
Cultivation of sisal in the Bahamas, 212-

224,

Cuscutaceae, 177.
Cycadaceae, 134.
Cyclospathe n. gen

14.

279

280

INDEX TO BAHAMAN PAPERS

Cyclbspathe northropi n. sp., 142, 143.
Cyperaceae, 136. .

Dasycladaceae, 130.

Dendrocygna, 64.

Dendroica, 53,

Dictyotaceae, 130..

Dioscoreacese, 146.

Diplactis, 105, 108, 109, iii, Fig.

Discosma, 104.

Domoecia, 74.

Doricha, 58.

Dromidia, 79.

Dryobates, 60.

Ebenaceae, 174.

Eriphia, 73,

Erythroxylon reticulatum n. sp., 159,

Plate 8.
Eupatorium bahamense n. sp., 188, Plate 19.
Euphorbiaceae, 162-164.
Eutheia, 56.

Falco, 60.
Ferns, 133, 134.

Flora of New Providence and Andros, 119-
211.

Analysis of collection, 127-128.

• Botanical regions, 123-126, 211.

Endemic species, 207.

Notes on distribution, 207, 211.

Relations of flora, 190-193.

Tables showing distribution, 194-207.

Fregata, 64.
Fucaceae, 130.
Fungi, list, 132.

Galeoscoptes, 52.
Gecarcinus, 71, 72.
Gemmaria, 104.
Gentianaceae, 175.
Geology of Bahamas, 29-47.

Banana-holes, 17, 36, 38.

BoiHng holes, 35, 36.

Caves, 8, 33, 34.

Erosion, 31, 32.

Effects of vegetation on surface rhizo-

morphs, 38, 41.

Reefs, 13, 44.

Subsidence or elevation, 42-45.

Surface deposits, 31.

Geothlypis, 54.
Goniopsis, 73.
Gonodactylus, 95.
Goodeniaceae, 188.
Gramineae, 135.
Grapsus, 73.

Haematopus, 61.

Helicteres spiralis n. sp., 167, 168, Plate.
Heliotropium nanum n. sp., 178, Plate.
Helminthocladiaceae, 131.

Heteractaea, 75.

Heteractis lucida, 103-107, 117, Fig.

Himantopus, 61.

Hydrophyllaceae, 177.

HymenocaUis arencola n. sp., 146, Plate 2,

Hypericaceae, 168.

Icturus northropi n. sp., 56, 66-68. Col-
ored plate.
Ilicaceae, 165.
Isaurus, 104, 11 2-1 15, Fig.

Labiatae, 180.

Land crabs, 8, 18.

Larus, 64.

Lauraceae, 153, 154.

Leander northropi n. sp., 87, 88, Fig.

Lebrunea, 104, 108.

Leiolephus, 72.

Lentibulariaceae, 182.

Lichens, Ust, 132.

Liliaceae, 144.

Linaceae, 159.

Linum bahamense n. sp., 159, Plate 7.

Lithotrya, 95.

Loganiaceae, 174.

Lophactaea, 75.

Loranthaceae, 150.

Loxigilla, 6, 56.

Lythraceae, 170.

Macrocceloma, 77.

Malpighiaceae, 160,

Malvaceae, 167.

Mammals, 24, 48.

Map New Providence and Andros, 27.

Marine algae, 1 29-131.

Melastomaceae, 172.

Metastelma barbatum n. sp., 175, 176,

Plate 14.
Microphys, 77.
Mimocichla, 52.
Mimosaceae, 154, 155.
Mimus, 52.
Mithrax, 78, 79.
Mniotilta, 53.
Moringaceae, 164.
Mosses, list of, 133.
Myiarchis, 57.
Myrsinaceas, 173.
Myrstiphyllum ligustrifolium n. sp., 186,

Plate 17.
Myrtaceae, 170.

Naiadaceae, 135.

Nassau and environs, general description,

1-4.
New Providence, general description, 27, 119,

120.
Nyctaginaceae, 152.
Nycticorax, 63.

INDEX TO BAHAMAN PAPERS

281

Ocypodea, 70.
Oleaceae, 174.
Onagraceae, 172.
Orchidaceae, 146-149,
Othonia, 78.
Oxalidacese, 158.

Pachycheles, 80, 81.

Pachygrapsus, 73.

Palaemon, 87.

Palms, 136-144.

Palythoa, 104.

Panopeus, 74.

Panulirus, 83.

Papaveraceae, 154.

Papilionaceae, 157.

Passifloraceae, 169.

Paurotis n. gen., 138.

Paurotis androa, n. sp., 139, 140.

Pelecanus, 64.

Penaeus, 94.

Petrochirus, 82.

Petrolisthes, 81.

Phalacrocorax, 64.

Phoenicopterus, 12, 63, 64.

Phoradendron northropiae n. sp., 150,

Plate 4.
Phymanthus, 104.
Phytolaccaceae, 152.
Piranga, 56.
Pitangus, 57.
Pithecolobium bahamense n. sp., 155,

Plate 5.
Plagusia, 72,
Plumbaginaceae, 173.
Polioptila, 53.
Polygalaceae, 162.
Polygonaceae, 151.
Polypodiaceae, 133-134.
Porcellana, 80.
Portulacacese, 153.
Primulaceas, 173.
Pseudosquilla, 94.
Psilotaceae, 134.

Rallus, 62.

Ranunculaceae, 153.

Reptiles, 25.

Reynosia northropiana n. sp., 166, Plate

10.
Rhamnaceae, 166.
Rhizophoraceae, 170.
Rhodactis, 104.
Rhodomelaceae, 141.
Ricordea, 104, no.
Rosaceae, 154.
Rubiaceae, 184, 185.
Rutaceae, 161.

Sabalaceae, 137.
Samydaceae, 172.
Sapindaceae, 165.

Sapotaceae, 173.
Saurothera, 59.
Schizasaceae, 133.
Scrophulariaceae, 181.
Seiurus, 53.
Sesarma, 72.
Setophaga, 54.
Shells, list of, 99-102.

Gastropoda, marine, 100-102.

Gastropoda, terrestrial, 100.

Pelecypoda, 99.

Simarubeae, 161.

Smilacaceae, 145.

Solanaceae, 181.

Speotyto, 60.

Sphyrapicus, 60.

Spindalis, 6, 55.

Sporadinus, 59.

Starfishes, list of, 26.

Sterna, 64, 65.

Stenopus hispidus, 83, 84, Fig.

Stenopus scutellatus n. sp., 85, 86, Fig.

Sterculiaceae, 167.

Strix, 60.

Symphemia, 6r.

Tecoma bahamensis n. sp., 182, 183,

Plate 15.
Terminalia spinosa n. sp., 171, Plate 13.
Thrinax bahamensis n. sp., 137,, 138.
Tiliaceae, 167.
Tozeuma, 88.
Turneraceas, 169.
Typhaceaj, 135.
Tyrannus, 57.

Uca, 70, 71.
Umbelliferae, 173.

Valoniaceae, 130.

Vanilla articulata n. sp., 148, Plate 3.

Verbenaceae, 179.

Vireo, 55.

Vitaceae, 166.

Xylosma ilicifolia n. sp., 168, 169, Plate 12.

Zenaida, 61.

Zoanthus, 104, 110-112, Fig.

MISCELLANEOUS PAPERS

Eruption of Krakatoa, 225-239.

Plant Notes from Temiscouata County,

Canada, I, 240-249.
Plant Notes from Temiscouata County,

Canada, II, 250-254.
Plant Distribution of Mt. Washington,

N. H., 255-258.
Study of Histologj" of Stem of Hoya carnosa,

259-277.

QH Osborn, Henry Fairfield
109 A naturalist in the

B3 Bahamas
07

EioMea

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET

UNIVERSITY OF TORONTO LIBRARY

Mliiliiiilllllllllll!

II !

lilillliili^^^^^ I

llllil lilllllllllMillt

illillH

mill

nil!

Iiiiiii)ii|i|iiiiiiiiiii

UJiJlll JlJl

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "A naturalist in the Bahamas, John I. Northrop, October 12, 1861-June 25, 1891; a memorial volume"

See other formats