a !
a ;

a

t-1 !

a :-
a I

a

tr I
ru i

(V

MEMORIAL PAMPHLET

CONTAINING

CERTAIN DRAWINGS OF MEDUSAE

MADE BY

The late PROFESSOR WILLIAM KEITH BROOKS,
of the Johns Hopkins University.

MEMORIAL PAMPHLET

CONTAINING

CERTAIN DRAWINGS OF MEDUSAE

MADE BY

The late PROFESSOR WILLIAM KEITH BROOKS,
of the Johns Hopkins University.

NOTE.

This collection of drawings is extracted from the three volumes
of Medusae of the World, by Alfred G. Mayer, to whom they were
presented, for publication, either by Professor Brooks himself or by
the Department of Biology of the Johns Hopkins University, after
the death of Professor Brooks in 1908.

It is hoped that by gathering into one publication the beautifully
executed drawings of this faithful observer we may raise one more
tribute of respect to his memory as a leader in research.

DRAWINGS OF MEDUSAE

Made by the late Professor WILLIAM K. BROOKS.

Genus PODOCORYNE Sars, 1846.
Podocoryne carnea Sars.

SYNONYMS OF THE EUROPEAN FORM.

(?) D \smorphosa conchico/a, PHILIPPI, 1841, Archiv. fur Naturgesch., Jahrg. 8, Bd. i, p. 37, taf. I, fig. 3.

Podoearyne carnea, SARS, 1846, Fauna Littor. Norveg., p. 4, taf. I, figs. 7-18. — KROHN, 1851, Archiv. fur Naturgesch., Jahrg.
17, Bd. i, p. 266.— HINCKS, 1868, British Hydroid Zooph., p. 29, plate 5, 6 figs. — ALLMAN, 1871, Monograph Tubul. Hy-
droids, p. 349, plate 16, figs. 1-9. — DEVARENNE, 1882, Archiv. de Zool. Exper., tome 10, pp. 645, 674, 683, plate 33, figs.
6-15; plates 34, 36-38; 1882, Compt. rend. Paris, tome 94, p. 892. — WEISMANN, 1883, Sexualzellen bei den Hydromedu-
sen, pp. 63-72, taf. 19, fign. 1-13. — BROWNE, 1896, Proc. Zool. Soc. London, p. 463. — DE\TARENNE, 1881, Annals and
Mag. Nat. Hist., ser. 5, vol. 9, p. 134. — HAMANN, 1882, Jena. Zeitschrift fiir Natunvissen., Bd. I 5, p. 517, taf. 20, fign. I,
3, 4 (histology of hydroid). — JICKELI, 1883, Morphol. Jahrb., Bd. 8, p. 621, taf. 27 (histology of the hydroid). — GRAEFFE,
1884, Arbeit. Zool. Inst. Wien, Bd. 5, p. 347. — ISHIKAWA, 1888, Zeit. fur wissen. Zool., Bd. 47, p. 621, 6 fign. (origin of
egg cells). — BEDOT, 1905, Revue Suisse de Zool., tome 13, p. 103 (literature 1842-1850). — RITCHIE, 1907, Trans. Knv.
Soc. Edinburgh, vol. 45, p. 523 (from Cape Colony, South Africa).

Dysmorpkosa carnea, HAECKEL, 1879, Syst. der Medusen, p. 77.

Cytxh exigua, HAECKEL, Ibid., p. 634.

Podocoryne h.rckelii, HAMANN, 1882, Jena. Zeit. fur Naturwissen., Bd. 15, p. 519 (young stocks of P. carnea ?).

(?) Podocoryne conchicola (Philippi) in part, HARGITT, 1904, Mitth. Zool. Station Neapel, Bd. 16, p. 581, fig. 26, taf. 22.

SYNONYMS OF THE AMERICAN REPRESENTATIVE.

Turritopsis nutricula, AGASSIZ, A., 1862, Proc. Boston Soc. Nat. Hist., vol. 9, p. 97, figs. 22, 23; 1865, North Amer. Acal.,p. 167,
figs. 269, 270. — FEWKES, 1881, Bull. Mus. Comp. Zool., vol. 8, p. 153, plate 4, figs. 4, 7-10.

Calc\dion formosum, FEWKES, 1882, Bull. Mus. Comp. Zool., vol. 9, p. 294.

Podocoryne carnea, BUNTING, 1894, Journ. Morphol., vol. 9, p. 205, plate 1 1, fig. 68. — LEVINSEN, 1893, Vid. Meddel. Nat. For.
Kjobenhavn (5), Bd. 4, p. 153. — HARGITT, 1904, Bull. U. S. Bureau of Fisheries, vol. 24, p. 38, plate 4, fig. 5 (figure
labeled "Turritopsis nutricula," on p. 37).

Turritopsis nutricula (in part), NUTTING, 1901, Bull. U. S. Fish Commission for 1899, vol. 19, p. 375.

Porocoryne carnea, HARGITT, 1901, American Naturalist, vol. 35, p. 582, fig. 44.

Podocoryne, HAZEN, 1902, Amer. Naturalist, vol. 36, p. 193 (regeneration).

AMERICAN VARIETY.

Adult medusa. — The bell is ellipsoidal in form and is about 3.5 mm. in height. Gelat-
inous substance not very thick, but quite tough and rigid. There are about 24 to 32 mar-
ginal tentacles which are about as long as the bell-height. The tentacles are not very flexible
and are usually carried curled upward. The tentacle-bulbs are well developed and are filled
with entodermal pigment granules. The velum is well developed. There are 4 straight
and narrow radial-canals. The manubnum is flask-shaped and there is no peduncle. The
mouth-opening is surrounded by 4 short, radially situated, unbranched, oral tentacles, each of
which terminates in a knob-shaped cluster of nematocysts. The ripe ova and spermatozoa are
found in the 4 interradii within the ectoderm of the manubrium. According to Ishikawa and
Bunting, the ova originate in the entoderm, but the sperm originates and remains in the ecto-
derm. When the medusa is mature both ova and sperm are found in the ectoderm of the
stomach. In the female, the ova are large and prominent and are spherical in form. The
entoderm of the manubrium and of the tentacle-bulbs is red, or brown to red, in color.

Hydroid and young medusa. — No specific difference exists between the American and
European hydroid stocks of Podocoryne. The hydroid (plate 14, fig. 2) is commonly found
upon shells which are tenanted by the hermit-crab (Pagurus) and also upon the carapace of
Limulus. The polypites arise at somewhat irregular intervals from a hydrorhiza which clings
to the surface of the shell, etc., upon which the stock is growing. In young colonies the hydro-
rhiza consists of an open network of anastomosing fibers, which are covered by a thin, delicate
perisarc, and externally by a fleshy hydrocaulus, or coenosarc. As the colony becomes older,
however, the fibers of the hydrorhiza form a closer network, and the chitinous perisarc fills in

3

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Podocoryne carnea
Sars, of Europe.

Podocoryne carnea
var. americana.

Number of tentacles when liberated
from hvdroid.

4

4 to 8. Usually 8.

Number of tentacles possessed by
mature medusa.

Usually 4 to 8, rarely
16 (Graeffe).

16 to 32. Usually
about 24.

Medusa-buds on interradial sides of
stomach of medusa.

Observed bv Sars,
1846.

No medusa-buds
observed.

the spaces between them. Numerous short, chitmous spines are developed upon the crust
which covers the fibers, and thus we find the polypites arising from a flat, spinous base which
adheres to the surface on which the colony is growing. The colony is composed of two kinds
of hydranths: sterile feeding-polypites, and reproductive gonostyles. The sterile teeding-
polypites are spindle-shaped, being about twice as wide near the oral circlet of tentacles as
they are at the base. They have 12 to 1 6 straight, stiff tentacles. The mouth is situated at the
apex of a dome-shaped proboscis. They are very contractile and may vary in length from about
5 to 15 mm. The reproductive polypites, or gonostyles, are frequently exactly similar in size
and shape to the feeding-polypites, and, in fact, are probably merely feeding-polypites which

have developed medusa-buds

Differences between European and American medusa of Podocoryne carnta. / ItT^fi •> ^1 Tn

other instances the gonostyles
are smaller and more slender,
and possess not more than 4
to 8 tentacles (see g' , fig. 2).
The medusa-buds arise from
a zone which is slightly below
the circlet of oral tentacles.
From 4 to 8 of these buds are
usually to be seen upon each
gonostyle. According to Mar-
tha Bunting, 1894, it appears
that the medusa-bud arises as an outpushing of both entoderm and ectoderm of the gonostyle.
As the bud progresses in its development, we find the ova in the entoderm of the manu-
brium. When a later stage has been reached they migrate from the entoderm into the
ectoderm. The spermatozoa, on the other hand, originate in the ectoderm of the inanu-
brium, as was shown by Weismann, 1883. When set free the medusa usually has 8 ten-
tacles: 4 radial and 4 interradial. The radial tentacles are usually better developed than the
interradial and in some individuals there are but 4 tentacles at the time of liberation, the inter-
radial ones not yet being developed.

It is remarkable that while in some stocks of Podocoryne the medusae are set free in an
immature state, in others the medusae are mature when liberated, the manubnum being
distended with sperm or ova, which are discharged almost immediately after the medusa is
set free. It is possible, as Allman, 1871, suggests, that this difference may be due to the influ-
ence of local conditions, which may be favorable in the one case and not so in the other to an
advanced development of the medusa. Krohn, 1851, and Loven, 1857, have observed stocks of
the European form of Podocoryne which were setting free mature medusa?. Indeed, we appear
to have a parallel case in Sarsia on the Massachusetts coast, where immature medusae are liber-
ated during the early spring months, whereas the medusae become ripe, discharge their genital
products, and wither upon the hydroid stock in May. Good figures of Podocoryne stocks which
are setting tree immature medusae have been given by Sars, 1846; Hincks, 1868; Allman, 1871;
etc. When set free the medusae commonly have 8 tentacles, 4 radial and 4 interradial. The
radial tentacles are usually more advanced than the interradial, and in some few individuals
there is no trace of interradial tentacles at the time when the medusa is liberated. The manu-
brium is short and fusiform and the mouth is surrounded by 4 radially situated, oral tentacles,
each of which terminates in a knob-shaped cluster of nematocysts. W'hen set free the medusa
is about 0.5 to 0.6 mm. in height.

In those medusae which are set tree in an immature condition there are at first 8 tentacles,
but these increase in number as growth proceeds and finally, when the medusa is about 3.5 mm.
in height, there are usually about 32 tentacles, 8 in each quadrant. The manubrium ot the
young, immature medusa is slender and fusiform, while in those medusae which are set tree in a
mature state it is globular and greatly distended with the genital products. The ectoderm ot
the hydroid is slightly bluish and translucent, while the entoderm is creamy-pink or silvery-
white in color.

This species has been found upon the Atlantic coasts of Europe, in the Mediterranean
Sea, and from Saldanha Bay, Cape Colony, South Africa. Levinsen, 1893, records it from the

OF MIOni'S.K MADE BY WILLIAM KK1TH BROOKS. 5

west coast of Greenland, and we have found it in great abundance in Narragansett and
Buzzard's Bays, on the southern coast of New England. It has not been taken at Beaufort,
North Carolina, nor at any station farther south. The medusae are very common in Narra-
gansett Bay from the middle of June until October, and I enjoyed exceptional opportunities

Fie. 74. — Podocoryne carnca, hydroid and young medusa. Drawn from life by Professor William K. Brooks,
and kindly presented to the author for publication in this work.

for observing their growth and development while studying at Dr. Alexander Agassiz's labo-
ratory at Newport, in 1892-96.

In 1881-82 de Varenne concluded that both eggs and sperm-cells originate in the entoderm
of the coenosarc of the hydroid and afterwards migrate into the medusa-bud. The more
careful researches of Weismann, 1883, however, refuted this view, showing that the male
germ-cells originate in the ectoderm ot the budding medusa, but do not wander from their
place of origin. On the other hand, the female germ-cells may possibly originate in the ecto-
derm; if so they soon wander into the entoderm of the budding gonophore, then into the spadix
of the medusa-bud and finally into the ectoderm of the manubrium of the medusa. According
to the later studies of Ishikawa and ot Bunting, however, the egg-cells originate in the ento-
derm of the medusa-bud and afterwards migrate into the ectoderm.

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Genus TURRITOPSIS McCrady, 1856.
Turritopsis nutricula McCrady.

Oceania (Turritopsis) nutricula, McCRADV, 1856, Proc. Elliott Soc. Charleston, pp. 1-36, plate 4, figs. l-io.

Turritopsis nutricula, McCRADY, 1857, Gymn. Charleston Harbor, p. 25, plate 8, fig. I. — BROOKS, 1883, Studies Johns Hopkins
Univ. Biol. Lab., vol. 2, p. 465. — BROOKS, 1886, Mem. Boston Soc. Nat. Hist., vol. 3, p. 388, plate 37. — HARGITT, 11)04,
Bull. U. S. Bureau of Fisheries, vol. 24, p. 37 (the figure is Podocoryne cornea). — RITTENHOUSE, 1907, Proc. Boston Soc.
Nat. Hist., vol. 33, p. 437, plates 30-35, figs. 1-55 (development).

\lodetria multitenlacttla, FEWKKS, iSSl, Bull. Mus. Comp. Zool. at Harvard College, vol. 8, p. 149, plate 3, figs. 7-9.

Modreria nutricula, FEWKES, 1 88;, Bull. Mus. Comp. Zool., vol. 9, p. 295.

non Turrilopsis nutricula, AT.ASSIZ, A., 1865, North Amer. Acal., p. 167, figs. 269, 270.

Turritopsis nutricula (in parl), HAECICEL, 1879, Syst. der Mcdusen, p. 66. — Ni rrixr.. 1901, Bull. U.S. Fish Comm.,vol. 19^.37?

EUROPEAN MEDUSA.

Oceania polycirrha, KEFERSTEIN, 1862, Zeit. fur wissen. Zool.. Bd. 12, p. 26, taf. 2, fign. 11-13.

Turritopsis [>ol\>irma, HAECKEL, 1879, Syst. der Medusen, p. 66.

Turritopsis poh'drrha, HARTLAUB, 189", \\':ss-:n. MeeresuntersucH. Kommis., Mem- K i> 1, Helgoland, Neue Folge, Bd. 2, p. 480.

taf. l6r, fig. 2.
(?) Cvttfis polystvld, WILL, 1844, Hone Terges'in:i-, p. 68, taf. z, fi^. v.

American medusa (plate 14, fig. 10). — Bell usually slightly pynform and about 4 to 5 mm. in
height. Bell-walls uniformly thin. There are 40 to ~o simple marginal tentacles, which are all
of about the same length and are somewhat shorter than the hell-height. These tentacles are
capable of much contraction or extension. Their basal bulbs are large and there is a single,
ectodermal pigment-spot upon the lower (centripetal) side of each tentacle near its place <;t
origin from the basal bulb. The surface of the tentacles is covered thickly with nematocyst-
cells. The velum is well developed. There are 4 straight, narrow radial-canals and a narrow
circular vessel. The manubrium is large and fills about half of the bell-cavity. The upper
part of the manubrium near the base consists of 4 radially situated masses ot large, highly
vacuolated, entodermal cells, through the midst of which the 4 radial-canals extend downward
into the stomach. These cells are indeed only the entodermal walls of the radial-canals (plate
14, fig. 13). The stomach is large and quadratic in cross-section. The cruciform mouth is
situated at the extremity of a short neck and is surrounded by a row of nematocyst-beanng
knobs (plate 15, fig. 12). The gonads are developed upon the sides of the stomach, where
they occur in the form ol a double, longitudinal, swollen region in each adradius. Their
outer surfaces are smooth. The entoderm of the stomach is dull-yellow or orange, or dull-
yellow streaked with orange. The tentacle-bulbs sometimes contain a little entodermal
orange pigment. The ocelli upon the tentacles are dark-brown or orange.

HydroiJ mid vonnij medusa. — The hydroid (fig. 76)
was found bv Brooks, 1883, on piles of a wharf at More-
head City, North Carolina. It is a Dendroila-^c. and
\ related t\< /). Johrnh Weisrnann. The stems of
the hydroid are from 8 to 10 mm. in height and hear
large, terminal hydranths. There are also numbers ot
short, side branches which terminate in hydranths. The
mam stem and the side branches are incased in a loose,
cylindrical perisarc, which is thick and becomes incrusted
with foreign matter. The perisarc is not annulated, and
terminates abruptly by a sharp collar immediately below
each hydranth. The hydranth or feeding-polypite is long
and fusiform and bears from 18 to 20 short, thick, fili-
form tentacles, which are arranged in three or more in-
definite rows or whorls. The medusa-buds originate
upon the sides of the stem at the bases of the hydranths.
Each medusa-bud is borne upon a short stalk or peduncle
and is closely invested by a thin capsule of perisarc.
When set free the young medusa has 8 tentacles. The manubrium is cone-shaped and there
is a large peduncle formed of highly-vacuolated cells. 4 prominent, nematocyst bearing knobs
surround the mouth. The hydranths are pale yellowish-red.

This medusa is found from the coast of Cuba to the southern coast of New England,
being about equally abundant in the northern and southern limits of its range. It is very
common in the Bahamas and at Tortugas, Florida. In Charleston Harbor, South Carolina.

FIG. 75. — Turritopsis nutricula, after Brooks, ill
Mem. Boston Soc. Natural Hi-r> r\ .

OF MKDUS.E MAUK HY WILLIAM KKITH HKouKS. i

it is commonly infested by larvae of Cunoctantha octonarta McCrady, although it appears to
be quite immune from this parasite in other places. I can detect no specific distinctions
between this medusa and " Turritopsis pol\'cirr/ni," which is occasionally seen off the Atlantic
coast of France and Germany. This form is well figured by Keferstein, 1862, and Hartlaub,
1897, who are the only European students who have observed the medusa on the eastern side
of the Atlantic.

W.HS:

Fic. 76. — Turritofisis nulnciila, after Brooks, in Mem. Boston Soc. Nat. Hist. Hydroid and young medusa.

Rittenhouse, 1907, has made an elaborate study of the development of T. nutncula. 1 he
ova develop in the ectoderm of the 4 interradial sides of the manubrium. The primitive ova
grow by the absorption of ovarian cells around them, as is common in other hydromedusae.
The yolk-spheres in the ovum are formed from the ovarian cells which it absorbs. About 20 to
35, rarely 50 or more, eggs are discharged into the water by the muscular rupture of the ovar-
ian walls between 5 and 6 o'clock in the morning. The discharged eggs are spherical, 0.116
mm. in diameter and have no membrane. They are yellowish-white, heavier than sea-water,
and opaque. The outer layer of finely-granular ectoplasm is distinct from the coarsely gran-
ular, yolk-laden endoplasm. Soon after being discharged the egg gives off two polar bodies
and is fertilized. Segmentation is total and approximately equal. The first two segmentation
planes are meridional and the third equatorial. The blastomeres remain quite tar apart,

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

FIG. loia. — ^ and B, Proboscidacl\la ornata, var. gemmifera showing a young stage with stolons bearing medusa buds.
C to F, young and mature stage of IVilhia brooksii sp. nov.; having 6 primary radial-canals, 24 terminal ramuli.
From Beaufort, North Carolina. Drawn from life, by Professor Brooks.

OF MEDUS/E MADE BY WILLIAM KKITH BROOKS.

9

touching only slightly. After this the cleavage becomes remarkably irregular, recalling the
extraordinary condition observed by Hargitt in Pennaria. A solid morula is formed, which has
no central segmentation cavity and which resembles a loosely-connected mass of irregularly
grouped cells rather than an embryo of any metazoan. The cleavages follow one after another
at intervals of 20 to 30 minutes.

As in Metschnikoff's Oceania annata and Margin's embryos of Pe nnaria, this irregularly
shaped morula gradually changes into an oval embryo, the surface of which becomes ciliated, so
that it swims upward from the bottom. This takes place at about 4 in the afternoon in eggs
which were laid between 5 and 6 in the morning.

Rittenhouse finds that during this period, when the loose, irregularly shaped mass of cells
shapes itself into an oval embryo, the ectoderm and entoderm are formed. During this period
"the cell-boundaries are lost for a short .time and a syncytium is formed. This syncytial
structure is crowded with yolk-granules and nuclei are scattered throughout the protoplasm.
The nuclei soon become more numerous near the periphery and the cell-walls begin to appear"
between the peripheral nuclei. These peripheral cells are to become the ectoderm, which is
soon separated from the inner, structureless mass by the development of the mesoglcea. This
inner mass afterwards acquires cell-boundaries between its nuclei, and still later a central
cavity, the coslenteric space, develops; and thus the entoderm is formed. This coelenteric
cavity does not develop, however, until the larva is 48 to 60 hours old. The syncytium con-
dition in Turritopsis is much more complete than is seen according to Hargitt in Pennaria, or
in Bougainvillia, according to Gerd, 1892.

Rittenhouse finds that during the earl)- stages of cell-division in Turritopsis the multipli-
cation is solely by mitosis, but that later, when the embryo becomes a mere irregularly arranged
mass of loosely-compacted cells, some of the nuclei divide amitotically.

When about 50 hours old the elongate, oval larva ceases to swim through the water and
settles down upon its side on the bottom. The larva then becomes the hydrorhiza, or root, of
the hydroid, and the first hydranth arises as a bud from about the middle of its length. The
tentacles of the hydranth develop in indefinite whorls, with 4 tentacles in each whorl, the
oldest tentacle being nearest to the mouth. In the mature hydroid the tentacles appear to
be irregularly scattered rather than being arranged in whorls.

Turritopsis should be reared under more natural conditions than those of the ordinary
laboratory in order to determine whether the remarkable, irregularly formed embryos described
by Rittenhouse be normal or merely the result of pathological states induced by adverse
conditions; but Miss Beckwith has recently shown that the cleavage of Pennaria is normally
irregular as is described by Hargitt.

Rittenhouse finds that when the embryo is in the loose-celled, morula stage it may be
divided into two masses, each one of which produces a normal planula larva of small size.

Genus WILLSIA Forbes, 1846.
Willsia brooksii sp. nov.

Beautiful drawings of a young stage and also of the adult condition of this medusa
were made by the late Prof. William K. Brooks, while he was at Beaufort, North Carolina,
and were found among his unpublished figures, after his death. They were kindly pre-
sented to me by the Department of Biology of Johns Hopkins University for publication in
this work, and it seems but fitting that the species should be named in honor of the great
naturalist who discovered it. It is closely allied and possibly identical with the European
Willsia stellata, although the 3-rayed center of the stomach appears to distinguish it.

In the young stage there are 6 simple, slender radial-canals, 60° apart. The bell-walls
are relatively thin and the bell somewhat higher than a hemisphere with a bluntly pointed
apex. The 6 tentacles are 5 to 6 times as long as the bell-diameter and have swollen,
nematocyst-bearing, outer extremities.

In the mature medusa the bell is flatter than a hemisphere, thick walled, with a shal-
low bell-cavity. Twenty-four tentacles alternate with 24 exumbrella, nematocyst tracts
each with several clusters of nettling cells. The manubnum has 6 lips. Stomach 3-rayed
at center, but each ray forks, giving 6 ramuli, 6 primary radial-canals which bifurcate twice,
giving 24 terminal branches. Gonads extend along sides of stomach. The size and color
can not be determined from Professor Brooks's drawings. Found at Beaufort, N. C.

10 MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Genus EUCHEILOTA McCrady, 1857.
Eucheilota duodecimalis var. parvum.
Dipleuron parvum, BROOKS, 1882, Studies Johns Hopkins Univ. Biol. Lab., vol. 2, p. 139.

This variety resembles E. JuoJt-cimalis in all respects excepting that the gonads are
developed upon only ^ of the radial-canals. The gonads are spherical in shape and found
upon 2 diametrically opposite radial-canals, near the circular vessel.

This form was found by Brooks at Beaufort, North Carolina, from June until August,
and in 1904 I found it in a surface-tow taken early in December, off Cape Fear, North Caro-
lina. It appears to be a local race or variety of E. duodecimalis.

FIG. icia. — A, Eucheilola duodecimalis; B, c and D, Eucheilota duodecimalis var. parvum. Drawn by the late Prof. William K.
Brooks at Beaufort, North Carolina, and presented by the Department of Biology of Johns Hopkins University for pub-
lication in this work.

OF MEDVS.E MADE BY WILLIAM KEITH BROOKS. 11

Genus DIPLEUROSOMA Axel Boeck, 1866, sens, emend.
Dipleurosoma brooksii sp. nov.

This form is described from drawings made by the late Prof. William K. Brooks, and
tound after his death among his unpublished figures. The drawings were generously pre-
sented to the author by the Department of Biology of the Johns Hopkins University for
reproduction in this work.

Bell flatter than a hemisphere, evenly rounded, thin walled. Manubrium small, short,
and 8-sided. 8 simple lips, 4 long and 4 short. 8 radial-canals arise from the stomach;
the 4 in the radii of the long lips are trident-shaped, each giving rise to 2 side branches.
The 4 canals in the radii of the short lips are simple. Thus 16 terminal branches reach
the circular vessel. In the specimen figured by Professor Brooks only 14 canals joined with

FIG. Ii8a. — Dipleurosoma brooksii sp. nov. A, oral view; B, side view
with tentacles cut off short; c, oral view of lips.
Drawn from life by Professor Brooks.

the circular vessel, for two of the side branches of the main radial-canals failed to develop.
There appear to be typically 16 tentacles, one at the base of each terminal branch of the
radial-canals, although in two intervals there are small additional tentacles thus giving 18
tentacles in the specimen figured by Professor Brooks. The tentacles have long, conical,
tapering basal bulbs, which are hollow and very long, lash-like shafts, longer than the bell-
diameter and very flexible. No ocelli are figured by Professor Brooks. The gonads are
upon the sides of the 16 radial-canals, not touching the ring-canal but extending from the
sides of the stomach outward. The velum is a wide, annular diaphragm. The size and
color of the medusa can not be determined from Professor Brooks's drawings. The
medusa was found at Nassau. Bahama Islands.

12

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Were it not for the absence of ocelli I would be inclined to regard this medusa as a
hypertrophic specimen of Staurodiscus tetrastaurus in which the side branches of the 4
primary radial-canals had reached the circular vessel and 4 intermediate canals had
developed. I have, however, never observed such a condition among many hundreds of
apparently mature medusae of S. tetrastaurus found at Tortugas, Florida ; and it seems
probable therefore that Professor Brooks's medusa is a new species which may have been
derived from a Staurodiscus-like ancestor.

FIG. 160. — Eutima mira, after Brooks, in Mem. Boston Soc. Nat. Hist., 1886. Small figures are of natural size

and illustrate attitudes of the medusa.

OF MEDUSA MADE BY WILLIAM KEITH BROOKS. 13

Genus EUTIMA McCrady, 1857.
Eutima mira McCrady.

Eutima mira, MCCRADY, 1857, Gymn. Charleston Harbor, p. 88, plate n, figs. 8, 9. — AGASSIZ, L., 1862, Cont. Nat. Hist.
U. S., vol. 4, p. 363.— AGASSIZ, A., 1865, North Amer. Acaleplue, p. 116.— HAECKEL, 1879, Syst. der Medusen, p. 191.—
BROOKS, 1884, Zool. Anzeigcr, Jahrg. 7, p. 709; 1886, Mem. Boston Soc. Nat. Hist., vol. 3, p. 395, plates 38, 39
(hydroid). — NUTTING, 1901, Bull. U. S. Fish Commission, vol. 19, p. 378, fig. 93.

Eutima limpida, AGASSIZ, A., 1865, North American Acal., p. 116, figs. 173, 178. •

Eutima etnarginata (young medusa), BROOKS, 1882, Studies Johns Hopkins Biol. Lab., vol. 2, p. 141.

Eutima mira—E. limpida, HARGITT, 1908, Biol. Bulletin, vol. 14, p. ill.

Eutima gracilis (young medusa), FEWKF.S, 1881, Bull. Museum Comp. Zool. at Harvard College, vol. 8, p. 158, plate 5, figs. 1-4.

Bell about 1.5 times as broad as high, about 15 to 30 mm. in diameter. There are 4
radially situated tentacles, each about 3 times as long as bell-diameter; also about 100 small,
rudimentary nodules upon the bell-margin, some of which give rise to cirri. There are usually
no lateral cirri flanking the tentacles when adult, but these commonly occur in the young
medusa. There are 8 lithocysts, 2 in each quadrant. Each lithocyst contains 4 to 8 spherical,
highly refractive concretions. Velum quite well developed. There are 4 straight, slender radial-

Fir.. 161. — Larva of Euti

ma mira

, after Brooks, in Mem. Boston Soc. Nat. Hist., vol. 3.

canals and a narrow circular vessel. The peduncle is about 3 times as long as height of bell,
long, conical, and tapering gradually throughout its length from inner apex of bell-cavity to
stomach. Stomach small and flask-shaped, its proximal part, near point of union with
peduncle, thrown into complex folds. There are 4 simple, slightly recurved lips. The gonads
are situated upon the radial tubes, in two separate regions, one upon the peduncle, and one
upon the subumbrella.

The stomach, gonads, and tentacles, opaque blue-white in color. Many specimens display
green entodermal pigment in the stomach and in the basal bulbs of the 4 long tentacles.

The medusa is common at the Tortugas, Florida; Charleston, South Carolina; and
Beaufort, North Carolina. It is an occasional visitant to Newport, Rhode Island, and to

14 MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Woods Hole, Massachusetts, late in summer, being abundant in some years and rare in others.
. I am in accord with Hargitt in believing that E. rnira is identical with E. hmpida.

The development of Eutiina mira has been studied by Brooks, 1884 and 1886, who
reared the hydroid from the egg. The gastmla is formed by delamination of the entoderm
from the inner ends of the ectoderm cells. This takes place most rapidly at the narrow end
of the pear-shaped planula. This narrow end afterwards becomes invaginated; the invag-
inated cells are, however, ectodermal, and have nothing to do with the gastrula cavity, but
they form the cement gland which soon serves to attach the larva. One lip of this orifice ot
invagination grows faster than the other, so that the cavity is soon pushed to one side and
the larva becomes bilateral. When the larva attaches itself this invaginated ectoderm is pro-
truded and pours out its cement. The planula then elongates and forms a layer of perisarc
which fastens it throughout its entire length. It thus becomes a hydrorhiza, not a hydranth.
The first hydranth buds out at right angles to the length of the hydrorhiza at the end opposite
to the cement gland. As soon as the first hydranth has acquired mouth and tentacles another
buds out close to the base of the first and so on. The planula, therefore, persists as a root
(hydrorhiza) and produces the hydranths by budding.

The young hydranth has a tentacular basal web and 5 large alternating with 5 smaller
tentacles; 10 in all. The body of the hydranth is elongated and cylindrical, and is not covered
by the perisarc which invests the unannulated stem. The hydroid is a Campanopsis, very
similar to that from which Claus, 1881, reared Eutima (Octorclns] gcgcnbaitn.

Eutima cuculata Brooks.
Eutima cuculata, BROOKS, 1883, Studies Johns Hopkins Univ. Biol. Lab., vol. z, p. 140.

Bell about 8 mm. in diameter and quite flat, being about 4 times as wide as high. The
gelatinous substance is quite thick in the center, so that the cavity of the bell is very shallow
and forms less than half of the total height of the bell. The gelatinous substance diminishes
gradually in thickness from the center of the margin, where it forms a thin edge. 4 slender
tentacles, which are 3 or 4 times as long as bell-diameter. The basal bulbs of these tentacles
are small and there are no lateral cirri. Above the base of each tentacle is a small semi-
circular flap or hood, which is formed from the gelatinous substance of the bell. There are 9
or 10 very small marginal cirri in each quadrant. There are 8 lithocysts, 2 in each quadrant.
Each lithocyst contains 3 to 8 spherical concretions. There are 4 slender radial-tubes. The
peduncle is about as long as the bell diameter. It is conical above and prismatic below.
Stomach about one-fourth as long as peduncle, with 4 simple lips. There are 4 gonads upon
the 4 radial-canals, one on each canal. They extend from near the bell-margin to base ot
peduncle, but do not run down along the prismatic part of the peduncle.

The stomach and the entoderm of the tentacle-bulbs are intense green by reflected light.

This species was found by Professor Brooks at Beaufort, North Carolina, on August
7, 1880.

The bell is flatter than in Eutima mira McCrady, the tentacles are longer, and the
marginal cirri less numerous, but the only really distinctive feature appears to be the semi-
circular flaps above each tentacle-bulb.

OF MKDUS.K MA11F, BY WILLIAM KEITH BKooKS.

15

Genus PERSA McCrady, 1857.
Persa incolorata McCrady.

Pena incolorata, MCCRADY, 1857, Gymn. Charleston Harbor, p. 104, plate 12, fig. 3. — AGASSIZ, A., 1865, North Amer. Acal.,
p. 50. — HAECKEL, 1879, Syst. der Meduscn, p. 279.

Bell oval and about 3 mm. in height and 1.5 mm. in diameter; walls thin. There area
large number of long, brittle, club-shaped, ciliated tentacles. 8 sensory-clubs between the 8
radial tubes. Velum large and muscular. 8 very narrow, straight radial tubes and a slender
circular vessel. Peduncle short and conical. Stomach cylindrical, about half as long as bell-
height, with 4 long cruciform lips. The 2 gonads are developed on 2 diametrically opposed radial
tubes. They arise from these tubes upon the subumbrella and project inward into the
bell-cavity.

FIG. 261. — Persa incolorata, young medusa, from Beaufort, North Carolina.
FIG. 262. — Persa incolorata, mature medusa, from Beaufort, North Carolina.
Drawn from life, by Prof. W. K. Brooks.

16

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

The animal is transparent, excepting for the gonads, which are pale-yellow.

This rare species was found by McCrady in 1857 in Charleston Harbor, South Carolina,
and Brooks has found it in abundance during the summer at Beaufort, North Carolina. I
believe it to be identical with the Mediterranean Persa.

The figures of the medusa here presented were drawn from nature by the late Professor
William K. Brooks, who kindly presented them to me for use in this work.

Genus CUNOCTANTHA Haeckel, 1879, sens emend.
Cunoctantha octonaria Haeckel.

Cunina octonaria, MrCRADY, 1857, Gymn. Charleston Harbor, p. 109, plate iz, figs. 4, 5; Proc. Elliott Soc. Nat. Hist. Charles-
ton, S. C., vol. i, plates 4-7.— AGASSIZ, L., 1861, Cont. Nat. Hist. U. S., vol. 4, p. 168.— BROOKS, 1883, Studies Biol.
Lab. Johns Hopkins Univ., vol. 2, p. 467; 1884, Zool. Anzeiger, Jahrg. 7, p. 710; 1886, Johns Hopkins Univ. Circular,
No. 49, p. 86.— MAAS, 1893, Ergeb. der Plankton Exped., Bd. z, K. c., p. 53.— AGASSIZ, A., and MAYER, 1899, Bull.
Mus. Comp. Zool. at Harvard College, vol. 31, p. 166.

Foreo/ia octonaria, AGASSIZ, A., 1865, North Amer. Acal., p. 51.

Cunoctantha octonaria, HAECKEL, 1879, Syst. der Medusen, p. 316.— BROOKS, 1886, Mem. Boston Soc. Nat. Hist., vol. 3, p.
361, plates 43, 44. — WILSON, H. V., 1887, Studies Biol. Lab. Johns Hopkins Univ., vol. 4, p. 95, plates 1-3, 19 figs. — MAAS,
iSgz.Zoolog. Jahrb. Anat.Abth., Bd. 5, Heft 2, p. 274.— BIGELOW, H. B., 1909, Mem. Mus. Comp. Zool. at Harvard
College, vol. 37, p. 52, plates 14, 15, and 17.

Mature medusa (plate 55, fig. i). — Bell about " mm. in diameter, flatter than a hemi-
sphere. 8 tapering tentacles, somewhat longer than bell-radius, project from sides of bell
about midway between margin and apex; these tentacles are stiff and capable of but little
movement and are carried bent downward in scimitar-like curves. The entodermal core of
each tentacle consists of a row of disk-shaped, vacuolated cells. This core projects inward into
the gelatinous substance of the exumbrella, immediately above the middle of the outer edge of
one of the 8 radial stomach-pouches. The root of each tentacle, where it fits into the
gelatinous substance of the exumbrella, is thus covered with two -layers of ectoderm, one being

FIG. 305. — Larvae of Cunoctanlha attached to mouth of Turriiopsis nutricula.
From drawing by Prof. W. K. Brooks.

continuous with the general ectoderm covering the exumbrella and the other is continuous
with the ectoderm of the main shaft of the tentacle. The ectoderm of the shaft of each tentacle
contains many nematocysts and also deep-lying muscle fibers. Just beneath and centrifugal
to each tentacle-root is a thick pad of ectoderm-cells. This is the "peronium" of Haeckel
and it probably serves to support the tentacle that arises immediately above it. 8 loops of

OF MEDUS/E MADE BY WILLIAM KEITH BROOKS.

17

FIG. 304. — Cunoctantha octonaria, after Brooks, in Mem. Boston Soc. Nat. Hist., vol. 3. Showing development of actinula

larva and young medusa.

At young actinula. B and C, budding actinula? giving rise to medusa-buds. D, young medusa with only 4 tentacles.

E, aboral view of young medusa.

18 MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

the nerve-cord extend outward from the 8 peronia to the sense-organs upon the bell-margin.
These 8 loops are put into connection one with another by means ot nerve tracts that run
through the peronia. Thus the nervous system consists of an 8-lobed, star-shaped ring ot
ciliated ectodermal cells situated upon the exumbrella (see plate 55, figs. I and 2, n, «)•

In the adult medusa there are typically 24 marginal sense-organs, 3 in each octant of the
bell, although occasionally there are as few as 2 or as many as 5 in an octant. Each sensory
club is situated upon a small, hemispherical elevation of the bell-margin called the sensory
cushion. The sensory cushion is formed from the cells of the nerve-cord, which at this place
are elongate and spindle-shaped and bear long delicate bristles. Centripetal to the sensory
cushion, on the exumbrella surface of the bell, there is a small, elliptical, elevated ridge, the
otoporpa, composed of a single layer of cube-shaped ectoderm-cells and containing a number
of nematocys'.s (see plate 55, figs. I and 2, o). The sensory-club is rod-like and cylindrical
and is attached to the sensory cushion by means of a short, narrow neck (see plate 55, fig. i')-
Each sensory-club contains two concretions that are situated within the entodermal core of
the club. The proximal otolith is small and spherical, while the distal is large, flat, and
crystalline.

The velum is more complex than in most hydromedusae, for it consists not only oi a simple
annular membrane (see plate 55, figs. I and 2, v) that serves to partially close the opening of
the bell-cavity; but in addition it extends upward as 8 A-shaped webs between the 8 loops of
the nerve-cord (see plate 55, figs. I and 2 i1'), and thus it is that, morphologically speaking, the
margin of the exumbrella is bounded by the nerve-cord and hence the true form of the disk is
to be considered as an 8-rayed star, although the thin webs of the velum stretched between the
rays give it the appearance of a hemispherical bell.

Manubrium cone-shaped and there are 4 perradial, cruciform lips. 8 gastric sacs radiate
outward from the central stomach-cavity, each sac being found in a tentacular radius. There
are no radial-canals and no ring-canals in this species. The gonads are developed in the ecto-
derm of the subumbrella immediately beneath the 8 gastric sacs.

The medusa is colorless with the exception of a delicate sage-green tinge of the entoderm
near the lips and occasionally in the tentacles.

Genus PELAGIA Pe"ron and Lesueur, 1809
Pelagia noctiluca P^ron and Lesueur.

Medusa ao.'tiluca, FORSKAL, 1775. IV-mpt anim. itin. orient., p. 109.

Pelagia r.o-nlura, PERI N u I.i M EUR, iSoq. Anna!, ilu Mus. Hist. Nat., turn.- 14. p. ;<;o; /'. furf,u,.i, Aurcllia pkosphorira,
I'-. . ,it.. pp. -,50, 358.— KROHN. 1855, Mullen's Archiv. Anat. Physio!., p. 491, taf. 20 (development).— HAECKEL, 1880,
s\ t. der Mc-d.is-n.p. 505 (list of authors and names).— KOWALKVSKV, iS-;, Mem. Imp. Soc. Lovers of Nat. Hist.. Moscow,
vol. io. pan :. p. -. plate 3 (development).— HAMANX, 1883, Zeit. fur \vissen. Zo..l., Bd. 38, p. 422, taf. 32 (development
and structure of gonads).— METSCHNIKOFF, 1886, Embrvol. Studien an Medusen, \Vi.-n., p. 24 (egg); 67 (segmentation);
ico (larva); taf. 10, fu;n. 23-28.— MONACO, Prince of, 1887, Comp. Rend. Paris, tome 104, p. 452 (swarming habits of
the medusa).— V^XKOFT FN-, iSSS, Bibliotheca Zoologica, Heft. 3, p. 8, taf. I, fi.cn. 5, 6; taf. 6, fign. 1-5; 1908, Deutcch.
Sudpclar Expedition, 1901-1903, Bd. io, Zool. 2, p. 38.— GOETTE. 1893, Zeit. fur wissen. Zoo]., Bd. 55, p. (-59. n fign.,
taf. 30-31; 189}, Sirzunsslfcr. Ak.id. Wissen. Berlin, p. 853 (development).— SCHAXEL, 1910, Zool. Anzeigcr, BJ. 35, p.
407 (histology of cbgcnesis).

The following is a description of a typical, adult specimen from the Bay of Naples:
Disk somewhat higher than a hemisphere when contracted, but flatter than a hemisphere
when expanded. In ordinary contraction it is about 49 to 55 mm. in diameter and 31 mm.
high. Sides of bell relatively straight and sloping, the apex flat. Numerous nettle-wans
over the exumbrella, arranged in more or less irregular lines radiating trom aboral apex of
exumbrella. These warts are rich orange-red in color and are elongate and linear, some-
times with, but more often without, cross-foldings. Near the bell-margin, however, they
lose their linear shape and become small, simple, and more or less oval.

The 8 marginal sense-organs are set in deep niches in the perradii and interradii. The
sense-club has no ocellus, but contains only a terminal mass of deeply pigmented orange-colored
crystalline concretions of entodermal origin. There is no sensory pit in the exumbrella above
the sense-club. The 8 hollow, tapering tentacles are each about twice (115 mm.) as long as
the bell-diameter.

OF MEDUSAE MADE BY WILLIAM KEITH BROOKS. 19

There are 16 subrectangular marginal lappets, with shallow median notches and rounded
angles. The septum between the ultimate branches of the radiating stomach-pouches in the
marginal lappets is twice as wide as the ultimate pouches themselves. The 4-sided throat-
tube is as long as the bell-radius. The 4 lanceolate lips or palps, with their complexly tolded
margins are each about 1.33 as long as the bell-diameter. Thus in an adult medusa with a
disk 49 mm. wide the palps were 68 mm. long.

The bell has a rich rose-purple tinge; the gonads, the entodermal cores and the tentacles
being especially deep in this color. The warts upon the exumbrella and along the outer edges
of the palps are orange-brownish red.

This medusa is abundant at times in the Mediterranean, especially in summer, although
large specimens are rarely seen in winter. It may be locally abundant during several suc-
cessive seasons and then vanish for years. For many years it was all but unknown in the
Bay of Naples but since 1900 it has been one of the commonest Scyphomedusae in th;s region.
It ranges widely over the warm regions of the Atlantic.

The development has been studied by Krohn, Kowalevsky, Hamann, Goette, Hyde, and
Metschnikoff. Hamann, 1883, has made a detailed study of the development of the gonads,
and their structure has been described by the brothers Hertwig, 1878. They appear as 4
interradial, elongate ridges in the entoderm of the subumbrella. The entoderm forms a series
of follicles in which the sex-cells develop and then migrate into a gelatinous lamella between
the layers of entoderm.

According to Metschnikoff, the egg is violet-brown and is laid between 12 and 2 in the
afternoon, in December, in the Mediterranean. Segmentation is total and nearly equal, and
a very large central segmentation-cavity is formed. The gastrula results from invagmation
at the hinder end of the body. The blastopore does not close, but forms the mouth of the
larva. Thus, according to Goette, 1893, the mouth is ectodermal and forms by invagination
at the hinder end of the larva, but the invaginated sac by no means fills the segmentation
cavity. The first pair of stomach-pouches arise from the entoderm and are 180° apart, then
follows an ectodermal pair 90° away from the first. The latter then develop 2 lateral pouches
each, and at a later period the entodermal pair each gives rise to 2 lateral pouches, thus giving
a larva with 6 ectodermal and 6 entodermal stomach-pouches; finally the ectodermal pouches
give rise to 4 new adradial pouches and the larva has 16 stomach-pouches — 10 ectodermal
and 6 entodermal. There is thus a striking analogy between its development and that of the
scyphostoma of Aurelha, according to Goette.

The external features of the transformation of the free-swimming larva into the medusa
have been studied by Krohn (1855), Kowalevsky (1873), etc. The mouth-end of the larva
become expanded and crater-like, with the mouth at summit of central cone of crater. The
depressed region around the cone becomes the subumbrella. The lappets, into which the
gastrovascular cavity is continued, grow out at intervals around the margin. The covering
of cilia is lost from the body of the larva and it begins to swim by means of rhythmical con-
tractions of its oral disk. Thus the free-swimming scyphostoma is converted into a medusa
without sTobilization ("see Goette, 1893).

Reasoning by analogy from the excellent work of Hyde, 1894 (Zeit. fiir wissen. Zool., Bd.
58, p. 531), upon Aurellia, it is probable that only the subumbrella floor of the second pair
of evagmated gastric pouches is formed from ectoderm, their exumbrella sides being of
entoderm. (See also Hadzi's work upon Chrysaora.}

20 MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

Pelagia cyanella P£ron and Lesueur.

Medusa pelagica, LINNE, 1758, Systema Nature, Ed. 10, p. 660.

Medusa pelngica, LINNE, 1766, Systema Naturae, Ed. 12, p. 1098. — 1788 (Gmelin), tomus I, pars 6, p. 3154.

Pelagia cyanella, PERON ET LESUEUR, 1809, Ann. Mus. Hist. Nat. Paris, torn. 14, p. 349, No. 66.

Dianaea cyanella, LAMARCK, 1817, Hist. Anim. sans vert., tome 2, p. 507.

Pelagia cyanella, ESCHSCHOLTZ, 1829, Syst. der Acalephen, p. 75, taf. 6, fig. i. — Bosc, 1830, Hist. Nat. des Vers., Ed. 2, tome 2,
p. 140, plate 17, fig. 3. — AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 128, 164; Ibid., 1860, vol. 3, plate 12,
figs. 1-16. — AGASSIZ, A., 1865, North Amer. Acal., p. 47, fig. 68. — HAECKEL, 1880, Syst. der Medusen, p. 507. — VAN-
HOFFEN, 1888, Bibliotheca Zoologica, Bd. i, Heft. 3, p. 22. — BIGELOW, 1890, Johns Hopkins Univ. Circ., vol. 9, No. 80,
p. 66. — HARGITT, 1904, Bull. U. S. Bureau of Fisheries, vol. 24, p. 70, plate 7, fig. I.

This American medusa is very closely related to the European P. noctiluca, of which it
is apparently only a local variety.

Bell about 40 mm. high and 50 mm. broad; somewhat fuller than a hemisphere, being a
little less broad at margin than a short distance above. Numerous small wart-like nemato-

O

cyst capsules are sprinkled thickly over the exumbrella and are especially thick in a zone at
about mid-height of bell; these protuberances are reddish in color and tend to be arranged in
radiating lines. 8 very long, highly contractile, hollow tentacles alternate with 8 marginal
sense-organs. Each sense-club is set within a niche between two adjacent lappets and is
protected on the outer side by a partial web between the lappets. The club is hollow and has
no ocellus, but contains a terminal, entodermal mass of crystalline concretions which are deeply
pigmented. 1 6 marginal lappets, hemispherical in shape. There is a long, narrow, 4-sided
proboscis, the radial corners of which extend downward as 4 long, flexible mouth-arms, the
free edges of which are complexly crenulated. The proboscis, together with the mouth-arms,
or palps, is about 3 times as long as bell-height. There are 4 complexly folded horse-shoe-
shaped gonads in interradial positions upon the floor of the subumbrella, and immediately
centripetal to them are 4 subgenital pits or cavities extending inward from the outer surface
of the subumbrella. The quadrangular ossophagus leads into a circular, disk-shaped, central
stomach which gives rise to 16 radial pouches extending outward in the radii of the sense-
organs and tentacles. Each of these pouches gives off" a pair of unbranched, curved canals
which enter the lappets, but do not torm a ring-sinus. There are 16 well-developed strands ot
radiating muscle fibers in wall of exumbrella ad)acent to gastrovascular cavity. These extend
outward in the radii of the tentacles and sense-organs, and fork as they approach the bell-margin.

The color is quite variable, sometimes bluish, sometimes slightly yellowish. Exumbrella
and mouth-arms sprinkled over with brownish-red nettlmg-warts, tentacles reddish-purple.

This species is found among the West Indies and Florida Reefs, and in summer it may
drift northward in the Gulf Stream so as to appear off the southern coast of New England
from July to September.

L. Agassiz, 1860 and 1862, found that the planulae of this species, as in P. noctiluca,
develop directly into medusae without going through a sessile scyphostoma stage and without
alternations of generations. The planulae are set free into the water where each develops into
a single medusa. The minute details of the development have been worked out upon Pelagia
noctiluca by Metschnikoff, 1886 (Emb. Stud, an Medusen, Wien.), and by Goette, 1893
(Zeit. fur wissen. Zool., Bd. 55, pp. 659-692). The gastrula is formed by invagination. The
first pair of radial stomach-pouches appear, according to Goette, as outpocketings from the
entoderm and these are quickly followed by another pair from the ectoderm of the throat-
tube, the two latter being 90° away from the former. The ectodermal pouches then give rise
each to two side branches and soon thereafter the entodermal do the same. Thus the cen-
tral stomach comes to have 12 radial pouches. 4 more radial pouches are soon formed
from the ectodermal pouches, so that the young medusa finally possesses 16 radial pouches.
It follows in adult medusa that the center of the exumbrella side of the central stomach is derived
from entoderm. 2 diametrically opposed, perradial pouches are ectodermal in origin and the
other 2 are entodermal. The 4 interradial pouches are ectodermal, and of the 8 adradial
pouches, 4 are ectodermal and 4 entodermal. The wall of the oesophagus is of ectodermal
origin. The young medusa soon develops 8 lobes which bifurcate, giving 16 marginal lap-
pets. The 8 marginal sense-organs develop before the tentacles. The mouth is at first a sim-
ple, round opening at the center of the crater-like ectodermal depression. It soon acquires
4 lips, but the mouth-arms do not develop until a later stage. It is probable that the ecto-
derm does not take so large a share in the formation of the stomach-pouches as Goette sup-
poses (see Clirysaora and Aurellia).

BROOKS

Pclagia cyanclla, from the borders of the Gulf Stream off Woods Hole, Massachusetts.

Drawn from life, by the late 1'rnf. William K. Brooks

OF MKnrS.E MADE BY WILLIAM KEITH BROOKS. lM

Genus DACTYLOMETRA L. Agassiz, 1862.
Dactylometra quinquecirrha L. Agassiz.

Pelagla quinquecirrha, DESOR, E., 1848, Proc. Boston Soc. Nat. Hist., vol. 3, p. 76.

Dactylometra quinquecirrha, AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 125, 166. — AGASSIZ, A., 1865, North Amcr.
Acal., p. 48, fig. 69. — HAECKEL, 1880, Syst. der Medusen, p. 518. — FEWKES, 1882, Bull. Mus. Comp. Zoul. at Harvard
College, vol. 9, No. 8, p. 293, plate I, figs. 2^-28, 38, 39. — BIGELOW, 1890, Johns Hopkins Vniv. Circulars, vol. 9, No. 80,
p. 65. — AGAPSIZ, A., AND MAYER, 1898, Bull. Mus. Comp. Zool. at Harvard College, vol. 32, p. i, plates i-i i, 33 figs. —
HARGITT, 1904, Bull. U. S. Bureau of Fisheries, vol. 24, p. 69, plate 7, fig. 2. — HARGITT, 1905, Journal Exper. Zool., vol. 2,
p. 575 (variations). — VANHOFFEN, 1906, Nordischcs Plankton, Nr. n, p. 50, fign. 13-14.

Bathyluca Solaris (damaged and regenerating specimen r), MAYER, 1900, Bull. Mus. Comp. Zool. at Harvard College, vol. 37,
p. 2, plate i.

Chrysaora, BIGELOW, R.P., 1880, Johns Hopkins Univ. Circulars, vol. 9, No. 8, p. 66 (brackish-water variety from Chesapeake
Bay).

Adult medusa. — Bell nearly hemispherical, 170 to 190 mm. in diameter. Numerous
small, wart-like clusters of nematocysts thickly scattered over the exumbrella, especially
abundant at aboral apex where they appear as little hemispherical projections above the
general surface; near the margin they are elongate in shape, while at the margin itself they
are again hemispherical as at the apex. 8 marginal sense-organs, 40 tentacles, and 48 marginal
lappets. The marginal sense-organs are set within niches between the lappets, 4 being pcr-
radial in position and 4 interradial; these niches are protecied above by a small web between
the lappets. A ciliated, pit-like depression extends downward from the surface of the ex-
umbrella immediately above each sense-organ. The sensory-club projects slightly down-
ward and contains a distal, entodermal mass of crystalline concretions but no ocellus. The
entodermal core of the sense-club is hollow and its lumen is connected with the general
gastrovascular space of the medusa.

There are 5 tentacles between each successive pair of sense-organs. 3 of these tentacles,
the primary and secondary, arise from the clefts between the lappets, but the other 2 (tertiary)
are generally found to spring from the under or subumbrella side of the ocular lappets; for
even in very large medusa? the ocular lappets exhibit but a slight notch adjacent to the ter-
tiary tentacles; in fact, the tertiary tentacles do not usually make their appearance until
the medusa is about 130 mm. in diameter and the lappets remain undivided until the medusa
is mature, although Hargitt shows that this is subject to great individual variability. Thus
in immature medusas of large size there are usually but 24 tentacles and 32 marginal lappets,
and the animal is in the "Clirysaora stage." I believe, also, that they often mature in this
stage and never reach the Dactylometra condition.

The primary and secondary tentacles are very long and flexible while the tertiary ten-
tacles are only a few millimeters in length. In like manner the lappet-clefts of the primary
and secondary tentacles are deep and the lappets almost as long as they are broad; while
the lappet clefts of the tertiary tentacles are mere shallow notches in the contour of the lappets
adjacent to the sense-organs. Mouth-opening cruciform, in center of subumbrella, at extremity
of a 4-cornered oesophagus and surrounded by 4 mouth-arms or palps, which when fully
extended are about 7 or 4 times as long as the bell-diameter. The 8 free edges of the mouth-

•J r D O

arms are complexly crinkled and highly flexible. The central stomach occupies a wide lentic-
ular space in the midst of the bell and gives rise to 16 simple, radiating pockets, 8 in the
tentacular and 8 in the rhopalar radii. These pockets are completely separated one from
another by 16 radiating septa which join the upper and lower walls of the umbrella cavity
together. The tentacles are hollow throughout the greater part of their length and their
entoderm is ciliated as is that of the stomach itself.

The gonads are contained in 4 interradially situated, entodermal infoldings of the wall
of the subumbrella, and their position is marked by 4 deeply sunken, subgenital pits. The
genital organs are provided with numerous, simple, unbranched gastric cirri which project
inward into the stomach-cavity. There are two sets of radial muscle-fibers; the principal set
is found in the 16 septa between the gastric pouches, and alternating with these in position
are 16 strands in the exumbrella, 8 of which lead outward to the sense-organs and 8 to
the primary tentacles.

Color quite variable. In some individuals the disk is pink, in others yellow with a
bluish opalescence. The exumbrella is thickly sprinkled with yellow-ocher colored nettlmg-
warts and there are 16 radiating stripes of reddish color upon the exumbrella in the radii of

22

MEMORIAL PAMPHLET SHOWING CERTAIN DRAWINGS

the septa of the peripheral stomach. These reddish stripes extend about half-way from the
bell-margin toward the center of the exumbrella and are due to highly refractive, rosin-
colored pigment granules within the epithelial cells of the disk. The male gonads are
generally pink, while the ovaries are yellowish or ashy-gray. The radial muscle-strands of
the subumbrella are of a glistening white and the entodermal cores of the tentacles are pink.
The mouth-arms are pink or yellow and always sprinkled over with red-colored pigment spots.
The marginal sense-organs contain each a mass of glistening white concretions, but no ocelli.
This species extends from the southern coast of New England to the tropics. In August
it is abundant in Tampa Bay, Florida. It has been taken by Bickmore at the Bermudas,
and by Drayton between the Bermudas and the Azores. "A well-marked southern variety"

FIG. 372. — Young ephyra? of Dactylometra quinquecirrha. Figures drawn by the late Prof. William K. Brooks at the
Chesapeake Bay Laboratory of the Johns Hopkins University. Presented by the Department of Biol-
ogy of the Johns Hopkins University for publication in this work.

OP MEDUS/E MADE BY WILLIAM KEITH BHOUKS. 23

was found by Brooks at Beaufort, North Carolina, and is figured in plate 6^.A. It makes its
appearance upon the surface along the coast of New England in August when large medusae
are found. The young rarely come to view, but remain in deep water.

Varieties and development. — The egg develops into a free-swimming planula which
soon attaches itself to the bottom and develops into a scyphostoma having normally 4 ten-
tacles. R. P. Bigelow, 1880, states that the so-called "Chrysaora" of the Chesapeake, which
is only a brackish-water, abortive variety of Dactylometra, develops from an ephyra through a
Pelagia stage, wherein it has only 8 tentacles and 16 lappets, and Brooks has figured the ephyrae
in the text figures here shown.

The present writer found considerable numbers of Chrysaora-like medusae in Hampton
Roads and Norfolk Harbor, Virginia, and in St. Mary's River, Maryland, early in Novem-
ber, 1904 and 1905. These were generally pale milky-yellow in color and lacked the rich
brown pigment and the 16 pigmented, radial areas seen upon the exumbrella of Dactylome tra
quinquecirrha. Others had a red-brown spot at the apex of the exumbrella, and surrounding
this was a star-like zone of red-brown streaks with pointed ends directed outward. The
axial ribs of the mouth-arms (palps) were red-brown. Although all were in the Chrysaora
condition and had only 3 tentacles and 4 lappets in each octant, some appeared to be fully
mature or with gonads nearly ripe. The exumbrella surface and the palps were covered
with dull milky-yellow clusters of nematocysts. There were 8 marginal sense-organs as in
Dact\lometra, but only 24 tentacles and 32 marginal lappets. None of the medusas were,
however, as large as is commonly seen in full-grown Dactylometra quinquecirrha, the largest
Chrysaora-like medusa seen in Norfolk harbor being only 105 mm. in diameter. It should
be borne in mind that D. quinquecirrha does not usually attain 48 marginal lappets and 40
tentacles until the medusa is 120 mm. in diameter, and it seems therefore that the so-called
Chrysaora of the Chesapeake is only a stunted Dactylometra which becomes mature in the
Chrysaora stage, and its pale coloration may be a local peculiarity due to unfavorable con-
ditions of confinement in brackish water. In the purer ocean water at (he mouth of Chesapeake
Bay the medusns develop into the Dact\lometra condition with 40 tentacles. These conditions
are also found in Narragansett Bay, Rhode Island, where in relatively pure clean water the
medusae have 40 tentacles, but in brackish estuaries they often become mature with only 24
tentacles and are pale in color.