Dev Genes Evol (2006) 216:743–754

DOI 10.1007/s00427-006-0101-8

ORIGINAL ARTICLE

The evolution of colony-level development

in the Siphonophora (Cnidaria:Hydrozoa)
Casey W. Dunn & Günter P. Wagner

Received: 17 April 2006 / Accepted: 5 July 2006 / Published online: 16 September 2006
# Springer-Verlag 2006

Abstract Evolutionary developmental biology has focused other zooids from a single bud) is a synapomorphy of the
almost exclusively on multicellular organisms, but there are Codonophora. The origin of probud subdivision is associated
other relevant levels of biological organization that have with the origin of cormidia as integrated units of colony
remained largely neglected. Animal colonies are made up organization, and may have allowed for greater morpholog-
of multiple physiologically integrated and genetically ical and ecological diversification in the Codonophora
identical units called zooids that are each homologous to relative to the Cystonectae. It is also found that symmetry
solitary, free-living animals. Siphonophores, a group of is labile in siphonophores, with multiple gains and/or losses
pelagic hydrozoans (Cnidaria), have the most complex of directional asymmetry in the group. This descriptive work
colony-level organization of all animals. Here the colony- will enable future mechanistic and molecular studies of
level development of five siphonophore species, strategi- colony-level development in the siphonophores.
cally sampled across the siphonophore phylogeny, is
described from specimens collected using deep-sea sub- Keywords Major transition in evolution .
mersibles and by self-contained underwater breathing Asexual reproduction . Animal colonies . Division of labor .
apparatus diving. These species include three cystonects, Functional specialization
Bathyphysa sibogae, Rhizophysa filiformis, and Rhizophysa
eysenhardti, and two “physonects”, Agalma elegans and
Nanomia bijuga. These data, together with previous find- Introduction
ings, are analyzed in a phylogenetic framework to recon-
struct key features of the history of colony-level organization The siphonophores, a group of pelagic colonial hydrozoans
and development in the Siphonophora. It is shown that (Cnidaria), include the longest animals in the world
gonodendra and gastrozooids of the examined cystonects (Robison 1995) and are among the most abundant
arise as independent buds directly on the stem, whereas carnivores of the open ocean (Pugh 1984). Even so, they
probud subdivision (the origin of feeding, reproductive, and have largely escaped the public eye, and many biologists
are not aware of their existence. Siphonophores have not
always been so obscure. They were of central interest to
Communicated by M. Q. Martindale
zoologists of the nineteenth century because of another
C. W. Dunn (*) distinguishing feature—they are the most complex of all
Kewalo Marine Laboratory, University of Hawaii,
41 Ahui St.,
colonial animals. Colonial animals have a life cycle
Honolulu, HI 96813, USA wherein multiple asexually produced zooids, each of which
e-mail: cdunn@hawaii.edu is homologous to a free-living solitary animal, remain
attached and physiologically integrated throughout their
G. P. Wagner
lives. Siphonophores have both the highest degree of
Department Ecology and Evolutionary Biology, Yale University,
P.O. Box 208106, functional specialization between zooids and the greatest
New Haven, CT 06511, USA precision of colony-level organization of any group of
744 Dev Genes Evol (2006) 216:743–754

colonial animals (Beklemishev 1969; see Fig. 1 for a due to the difficulty of collecting them. Although they are
schematic of the organization of a siphonophore colony). often large and relatively abundant, siphonophores are also
There has been much interest in the origin of develop- extremely fragile, and most species are found only in the
mental mechanisms and organization at new hierarchical open ocean. There are published accounts of the early
levels of individuality (Buss 1987; Maynard Smith and embryology of several siphonophores (Carré and Carré
Szathmáry 1995). Most of this work has been concerned 1993; Gegenbaur 1853; Haeckel 1869), but these studies
with the transition from unicellularity to multicellularity stopped short of describing the origins and structure of the
(Bonner 2001) and includes important empirical studies on growth zones that are responsible for the colony-level
diverse multicellular organisms such as Volvox and Dic- development that continues throughout the life of a
tyostelium. It will not be possible, however, to know siphonophore. The colony-level development of only three
whether phenomena revealed by these studies are unique species has been described (Chun 1885; Dunn 2005;
to the transition from unicellularity to multicellularity or are Schneider 1896). These three Codonophora taxa share a
general properties of transitions in biological organization common mode of development called probud subdivision.
unless there is an empirical push to characterize develop- Probuds arise sequentially within the growth zone, which is
ment and morphology at other levels of biological also the site of stem elongation. Each cormidium, the well-
organization. These necessarily include coloniality, and defined group of multiple zooids that is reiterated along the
the extreme form of colonial organization found in siphono- length of the siphosomal stem (Fig. 1), arises from a probud
phores promises to be particularly informative. through a stereotypical series of subdivisions.
There have been few studies of the colony-level The ability to safely dive using self-contained underwa-
development and organization of siphonophores largely ter breathing apparatus (SCUBA) from ships in the open
ocean (Hamner 1975) and the refinement of midwater
sampling devices that can be mounted on submersible
vehicles (Youngbluth 1984) now make it possible to collect
Anterior many species of siphonophores intact for the first time.
Nectosomal Pneumatophore (float) These new tools have enabled us to make the first
Growth comparative study of siphonophore colony-level organiza-
Nectosome
Zone
tion and development that draws on taxa sampled across the
Nectophore full diversity of the group. In conjunction with a recent
molecular phylogeny (Dunn et al. 2005), this allows us to
reconstruct the history of major changes in colony-level
development and organization within siphonophores and to
begin to understand how a diversity of form has been
realized at this poorly understood level of biological
organization. This descriptive work also provides a foun-
dation for future molecular and mechanistic studies of
siphonophore development that are relevant to questions
Siphosomal
Growth
of general interest regarding the origins and evolution of
Siphosome
Zone individuality.
Bract

Gonodendron
(with gonophores) Materials and methods
Cormidium
Gastrozooid Fresh specimens were collected by manned submersibles,
remotely operated vehicles, blue-water diving, and land-
based diving. This material was prepared for scanning
electron microscopy (SEM) as previously described by
Ventral Posterior Dorsal Dunn (2005). Preserved siphonophores from Dr. Philip R.
Pugh’s collection (most of which had been fixed with 4%
Fig. 1 Schematic overview of siphonophore structure modified from formaldehyde in borax-buffered seawater) were also suc-
Dunn (2005). The depicted colony is a physonect and possesses a cessfully prepared for SEM despite the fact that some
pneumatophore (gas-filled float), nectosome (a region with necto-
phores, the propulsive zooids), and siphosome (a region with zooids
specimens were decades old. The same protocol was used
of mixed functional types). Cystonects lack a nectosome, whereas as for fresh material, beginning part way through with the
calycophorans lack a pneumatophore 500-mM NaCl wash.
Dev Genes Evol (2006) 216:743–754 745

The budding sequence of each species was inferred Table 1 Specimens examined
through static observations of the ontogenetic sequence of
Species Specimen Collection
developing zooids. The organization of zooids along a
mature portion of the stem was first established. The Bathyphysa sibogae BWP796-20-7 PRP
regularity of siphonophore organization made it possible Rhizophysa filiformis BWP1077-3 PRP
to then determine the identity of immature zooids further to BWP567-8 PRP
BWP566-17 PRP
the anterior based on their relative position.
BWP566-16 PRP
BWP588-21 PRP
BWP442 PRP
Results Rhizophysa eysenhardti BWP539-13 PRP
BWP1611-12 PRP
A note on terminology BWP634-6 PRP
BWP741-3 PRP
Haddock et al. (2005) consolidated and standardized the BWP778-12 PRP
BWP465 PRP
usage of terms for axes and orientation in siphonophores,
BWP465 PRP
and the present article adheres to this clarified scheme. The BWP802-1 PRP
nomenclature used to describe the functionally specialized BWP1086-6 PRP
zooids of siphonophores is similar to that used for the BWP814-12 PRP
Hydrozoa in general, but also includes several terms that Agalma elegans YPM 36365-36390 YPM
are specific to the group (Dunn 2005). The term gonoden- Nanomia bijuga YPM 36427-36444 YPM
dron is used here to describe any compound reproductive Apolemia sp. JSLII 1450-SS3 PRP
structure consisting of multiple zooids that are attached to Forskalia formosa YPM 36445-36448 YPM
Lychnagalma utricularia Tiburon 676-D6 (specimen
the stem via a common peduncle. The structure and zooid
destroyed)
composition of gonodendra vary widely across taxa; it is not
at all clear if they are homologous across siphonophores or The numbers given in YPM rows are museum catalog numbers. The
have independently arisen multiple times. specimens in PRP’s collection are identified by their mode
of collection and a unique identifier.
PRP Personal collection of Philip R. Pugh (National Oceanographic
Material examined Centre, UK); YPM the Yale Peabody Museum (New Haven,
CT); BWP blue-water plankton series; Tiburon the remotely operated
Specimen data for the examined material are listed in vehicle Tiburon (Monterey Bay Aquarium Research Institute);
JSL II the manned submersible Johnson Sealink II
Table 1. Specimens of Apolemia sp. and the cystonects are
(Harbor Branch Oceanographic Institute)
from Philip R. Pugh’s collection. Most Agalma elegans
material was collected in the bay at Villefranche-Sur-Mer,
France, in the months of March–May in 2003 and 2004; the
remaining A. elegans specimens were collected by blue- anterior end of the sequence (i.e., the youngest gastro-
water diving from the RV Oceanus in the summers of zooids) have no sign of gonodendra between them.
2000–2002 along the east coast of the USA. Nanomia Gonodendra arise further to the posterior as isolated buds
bijuga and Forskalia formosa were both collected in between and in line with the maturing gastrozooids.
Monterey Bay, CA, and the bay at Villefranche-Sur-Mer, The young gastrozooids of B. sibogae have pronounced
France. These newly acquired specimens have been lateral ridges known as ptera (Leloup 1936), which give the
deposited in the Yale Peabody Museum (catalog numbers gastrozooids a bract-like appearance. The striking similarity
can be found in Table 1). The Lychnagalma utricularia of these Bathyphysa gastrozooids to “physonect” bracts
specimen quickly deteriorated after being caught and was may indicate that physonect bracts are in fact modified
not preserved. polyps. Each young gastrozooid also has a lamella extend-
ing part way up its posterior side and further anchoring it to
Bathyphysa sibogae the stem, just as physonect bracts do. This holds the young
gastrozooid so that its distal end faces to the posterior.
The gastrozooids and gonodendra of Bathyphysa sibogae Although conspicuous in young gastrozooids, the ptera and
were found to be in a uniserial sequence (Fig. 2b). The lamellae are absent by the 30th gastrozooid in the examined
gastrozooids first appear as isolated buds at the anterior end specimen.
of the siphosome and grow a tentacle on their anterior side Leloup (1936) figured and described the early stages of
as they mature and are carried to the posterior by the development of Bathyphysa gonodendra. He did not,
elongating stem. The first several gastrozooids at the however, trace the fate of the young buds or describe their
746 Dev Genes Evol (2006) 216:743–754

Fig. 2 Schematic illustrations a b Uniserial c Triserial

of the budding sequences of
long-stemmed cystonects (the
Rhizophysidae). (b, c ventral
view) a Rhizophysa eysenhardti
(adapted from Kawamura 1910). A
b Uniserial budding, in which
the gastrozooids arise in a single R L
line, and the gonodendra are
later intercalated between them. P
The tentacles are borne on the
anterior side of the gastrozooids.
c Triserial budding, in which the Gastrozooid
gastrozooids arise in two outer Tentacle
rows, and the gonodendra are
found in a row between them. Gonodendron
The tentacles are borne on the Gastrozooid
side of the gastrozooid facing
the ventral midline Gonodendron

Bathyphysa sibogae Rhizophysa eysenhardti

Rhizophysa filiformis

later stages of development. The material examined here central style (i.e., the main axis of the gonodendron). A
was in good condition, and the full ontogenetic sequence of single bud first arises on the side of the branch (Fig. 3d).
gonodendron development was described (Fig. 3). The This bud takes on a distinctive shape and matures into a
youngest gonodendron buds are at first simple, smooth nectophore. The portion of the branch distal to the
evaginations that protrude from the stem (Fig. 3a). These nectophore matures into the palpon that terminates the
then elongate and become warty in appearance (Fig. 3b,c). branch. Gonophores then form from evaginations that arise
Each wart-like protuberance then elongates and develops along the branch (Fig. 3e), some of which arise distal to the
into a gonodendron branch (Fig. 3d–g). The branches do nectophore. Each branch bears on the order of seven to nine
not ramify further, and are all attached directly to the gonophores at maturity. The gonodendra of this species,

Fig. 3 Developmental sequence

N
of Bathyphysa sibogae gono- d
dendron. a–c Young gonoden- P
dra showing the origin of the N
side branches as evaginations of 0.5 mm
the main gonodendron axis. e
d–g Close-ups of isolated side a b c P
branches following the develop- Go
mental stage shown in (c).
h Photograph of preserved N 1 mm
gonodendron. Scale bar
(0.5 mm) applies to a–e, Scale f
bar (1 mm) applies to f, g; scale
bar (2 mm) applies to h. Go,
P
gonophore; N nectophore; Go
P gonopalpon h
g Go N

2 mm P
Dev Genes Evol (2006) 216:743–754 747

like those of the other Cystonectae, are thought to break few intermediate stages of cormidial development in each
away from the colony at maturity before releasing their colony. It also implies that zooids may be added to the
gametes (Totton 1965, P.R. Pugh, personal communication). colony at a very slow rate.
There is not a pronounced horn (sensu, Dunn 2005) in
Rhizophysa filiformis the siphosomal growth zone of mature specimens of A.
elegans, although the tip of the siphosomal growth zone
All examined specimens of Rhizophysa filiformis had the does form a slight overhang (Fig. 4d–j, marked “T” in each
same uniserial development of gastrozooids and gonoden- pane). Totton (1956) noted a well-developed horn (which
dra as B. sibogae (Fig. 2b). There were from one to six he called a nectostyle) in the larvae of this species, so it
gonodendra found between the mature gastrozooids. The seems to recede as the colony matures. The anteriormost-
gonodendra develop in the same manner as those of B. differentiated cormidium consists of six buds on a common
sibogae (Fig. 3) and have the same general structure at peduncle (as seen in cormidium 1 of the specimens figured
maturity. The gastrozooids of R. filiformis do not have in Fig. 4d–g and h–j). The largest of these is a bud that was
pronounced ptera at any stage. inferred to be the gastrozooid based on its rounded shape
and consistent position relative to the other buds. There are
Rhizophysa eysenhardti two buds on the anterior side of the gastrozooid base. The
proximal bud, which was identified at various stages of
The origin of the siphosomal elements of Rhizophysa development by its anteriormost position within the
eysenhardti is triserial (Fig. 2c). The buds of the two outer cormidia, differentiates into the primary male reproductive
rows give rise to gastrozooids. The gastrozooids are elements. The distal bud on the anterior side of the
staggered with respect to each other. The single tentacle gastrozooid peduncle differentiates into one, or perhaps
of each gastrozooid forms on the side facing the midline. several, gastric palpons. There is also a bud on the left side
When the gastrozooids are mature, it is not entirely obvious of the gastrozooid base at the same level as the bud of the
that they arose in two separate rows, but the alternate male reproductive elements. This lateral bud was inferred to
attachment of the tentacle to the left and right sides of the be a bract based on the differentiation of buds in the same
gastrozooids is clearly discernable. position in more posterior cormidia. Finally, there are two
The median row of siphosomal elements, found between buds to the posterior of each young gastrozooid. The one
the left and right rows of gastrozooids, consists solely of furthest to the posterior had a characteristic shape even
gonodendra. Gonodendron buds are found just as far to the early in development that indicated that it was a palpon,
anterior as the gastrozooid buds are. The gonodendron buds and its precocious development and location indicate that it
are not exactly in phase with the gastrozooid buds, and is the B-palpon. Between the B-palpon and the gastrozooid
there are a variable number of gonodendra between gastro- is a bud that in posterior cormidia differentiates into the
zooids at maturity (but always at least one). The gonodendra female reproductive elements. The B-palpon grows larger
of R. eysenhardti develop in the same manner as those of B. than all the other zooids of a cormidium by the point where
sibogae (Fig. 3). the cormidium is second in the sequence. Totton (1954)
considered the gastrozooid to be the posterior element of
Agalma elegans each cormidium. In light of the developmental sequence,
we differ with him on this point and consider the B-palpon
Our observations of the zooid organization of A. elegans to be the posteriormost structure.
(Fig. 4b,c) are consistent with Totton’s (1954, frontispiece— In addition to the bract that arises on the left of the base
reproduced here as Fig. 4a). Each cormidium consists, from of each gastrozooid, there is a lateral row of bracts on both
posterior to anterior, of a large palpon (marked as “B” in the left and right sides of the stem. These bracts are further
Totton’s (1954) frontispiece so it will be referred here as the from the ventral midline than any other zooids and are more
B-palpon), a female gonodendron with an associated developed than adjacent bracts arising from the base of the
palpon, a gastrozooid with multiple gastric palpons gastrozooids. No buds that could give rise to these lateral
attached to its peduncle, and several clusters of male bracts were ever observed within the derivatives of the
elements. Each cluster of male elements has a palpon, probud, and the lamellae of these bracts sometimes
several male gonophores, and bracts in various stages of extended along the stem slightly to the anterior of the tip
development. Bracts also are spread along the full length of the siphosomal growth zone. These findings suggest that
of the stem. Each cormidium can span up to 4 cm or more they arise directly on the side of the stem.
of the stem, and there are few developing cormidia within The male bud comes to be attached to the stem
the growth zone. This makes the inference of the budding independently from the gastrozooid, gastric palpon, and
sequence quite difficult in this species because there are gastrozooid-associated bract, and the female bud comes to
748 Dev Genes Evol (2006) 216:743–754

b d
a T Specimen 1
2
1
A 3 50 µm
V D 2 3
P
(Anterior)
e f A g

Cormidium
R L T
T T P
1 1
1 1
1 2 2
1 2
2 2
2
3
3
3 3 3 3

(Right) (Ventral) (Left)

1 mm

A
R L
c P h
T 1 Specimen 2
1 2
50 µm
2

(Anterior)
Cormidium

i R
A
L
j
T 1 T
P
Male 1
2 1 2
Bract
Gastric Palpon 2 2
3
Gastrozooid 3

Female
3
B Palpon
1 cm (Ventral) (Left)

Fig. 4 Agalma elegans—false coloring indicates zooid identity. Ante- stem relative to the other zooids. d–g SEM of the growth zone of
rior faces the top of the page, unless noted otherwise. a Left lateral view specimen 1, shown from four different views. h–j SEM of the growth
of the mature colony (adapted from Totton 1954). The bracts, which zone of specimen 2, shown from three different views. In d–j, the tip of
sheathe the siphosome, have not been colored. b Photographic collage the growth zone is labeled with a T, and the gastrozooids and B-palpons
showing the growth zone and young siphosomal stem of an anesthetized are numbered according to the cormidium they belong to, beginning
living specimen. Ventral view. Note that both the female gonodendron with the first differentiated cormidium. Gastrozooids 1–2 and B-palpon
and the female-associated palpon have been colored orange, and that 3 have been broken away (h–j). Additionally, gastrozooid 3 has been
the male gonophore and male associated bract and palpon buds are broken away (i). Where young gastrozooids were broken away, the buds
colored blue. Buds whose fate could not be assigned with confidence of the gastric palpons, and sometimes the left bracts, also broke off with
have been left uncolored. c Schematic of a mature cormidium. Ventral them. A Anterior; D dorsal; L left; P posterior; R right; V ventral
view. The lateral bracts are not shown in their actual position along the
Dev Genes Evol (2006) 216:743–754 749

be attached to the stem independently from the B-palpon anterior side of its peduncle. A palpon and two flanking
(this is most easily seen in Fig. 4i). Bracts later bud at the bracts are the first such zooids (Fig. 5e). Additional bracts
foot of the B-palpon. After the female bud separates from are added on both sides of the gastrozooid peduncle, with
the B-palpon, it takes on a bilobed shape (as in cormidium the youngest (i.e., the smallest) being the most distal
3 of Fig. 4i,j), and further to the posterior, one lobe was (Fig. 5f). Additional palpons bud on the anterior side of
seen to be the female-associated palpon and the other the the primary palpon, and all palpons initially share a
female gonodendron (not shown). These are attached to the common base and form a fanlike structure (Fig. 5g). The
stem independent from each other at maturity. Close to original palpon is the most distal in this structure, and the
the growth zone, there is only one male cluster per smallest (youngest) is the most proximal. A lateral bract
cormidium. In cormidia slightly to the posterior, it can be forms on each side of each palpon as it matures.
seen that a new male cluster arises directly on the stem, The zooids of successively more mature cormidia
separate from and to the anterior of the original one. This come to be spread out along the stem such that each
process repeats several times, with new male clusters being zooid is eventually attached to the stem independently
added anterior to the existing ones. (Fig. 5h,i). The bracts that arose on the gastrozooid
There are multiple gastric palpons attached to the peduncle come to be arranged in two lateral rows that
peduncle of gastrozooids in mature cormidia. It was not flank all the other zooids (the lamellae of the left row of
determined if these arise by subdivision of the single gastric bracts can be seen in Fig. 5i). The oldest (i.e., largest and
palpon bud that arise within the growth zone or if each most proximal) bract to arise on the gastrozooid
gastric palpon arises as an independent bud on the peduncle peduncle comes to be located furthest to the anterior,
of the gastrozooid. There are more gastric palpons in and the youngest (i.e., the smallest and most distal)
cormidia further to the posterior. remains the closest to the gastrozooid. As the palpons
Large specimens of A. elegans have an irregular spread out, the youngest (i.e., the smallest and most
organization of zooids in their posterior cormidia, but in proximal) comes to be located the furthest to the
all examined cases, the irregularity was consistent with anterior of the cormidium, and the first palpon to arise
zooid loss. In these specimens, cormidia close to the growth (i.e., the largest and most distal) remains closest to the
zone still have the regular organization originally described gastrozooid. The lateral bracts associated with each
by Totton (1954) and reconfirmed here. palpon move away from the ventral midline and come
The nectosomal growth zone of a single specimen was to be arranged in rows just inside of the rows of bracts
examined using SEM. It was found that each young that arose on the gastrozooid peduncle. Kawamura
nectophore had a small bud on the posterior side of its (1911) had already noted that the bracts of this species
peduncle. A similar bud has previously been found in the are arranged in two rows along each side of the stem.
nectosomal growth zone of Bargmannia elongata (Dunn The male and female gonodendra arise at the base
2005). of the palpons just inside of each palpon-associated
bract (Fig. 5i). Additional clusters, each consisting of a
Nanomia bijuga palpon, a male gonodendron, a female gonodendron, and
bracts, are added directly on the stem at the anterior end
Totton (1965, Fig. 35) figured the organization of zooids of each cormidium after all zooids have spread out.
within a cormidium of N. bijuga. He found that each Clusters may also sometimes be added between existing
cormidium consisted of a gastrozooid and a series of palpons.
palpons, all attached independently to the stem. A female There is a single tentacle on the anterior side of each
gonodendron and a cluster of male gonophores flanked gastrozooid. Likewise, each palpon has a palpacle (as the
each palpon and alternated sides from palpon to palpon. tentacle of a palpon is called) on its anterior side.
Totton noted that secondary palpons, also with male Tentacles and palpacles form as simple evaginations. The
and female structures at their base, were sometimes bud of the gastrozooid tentacle can be seen in Fig. 5f–h,
intercalated between the primary palpons in mature and the base of the tentacle, which has been broken away,
cormidia. can be seen in Fig. 5i. The palpacle rudiments can be seen
Our own observations of N. bijuga confirm Totton’s in Fig. 5i. Some palpons also have a large basal swelling on
earlier findings. In addition, we are also able to describe the their anterior side just distal to the palpacle.
budding sequence by which the zooids of this species arise.
There is a pronounced horn within the siphosomal growth Apolemia sp.
zone, and each cormidium arises as a simple probud close
to its tip (Fig. 5d). The bulk of the probud gives rise to the One specimen of Apolemia, belonging to the same
gastrozooid, with most of the other zooids arising on the undescribed species as “Apolemia 3” in the molecular
750 Dev Genes Evol (2006) 216:743–754

Fig. 5 Nanomia bijuga. False b

coloring indicates zooid identity.
Anterior faces the top of the a d
page in all panes. a Mature
colony (adapted from Kawamura
1911). The stem is twisted, but
the view is generally lateral.
b Photographic collage showing
the growth zone and young
cormidia of an anesthetized liv-
ing specimen. The stem is 50 µm
twisted such that the youngest
cormidia are shown from their e
left side, and the older (posteri-
or) cormidia are shown from
their right side. c Schematic of a
single mature cormidium. Ven-
tral view. The lateral bracts are

Cormidium
not shown in their actual posi- 50 µm
tion along the stem relative to
the other zooids. d–i SEMs of f
the youngest cormidia, shown
from their left side in an onto-
genetic sequence (from anterior 100 µm
to posterior). The largest bracts
have been removed, leaving
only their lamellae. d The three
youngest cormidia at the tip of 1 mm g
the horn; all other panes show
only one cormidium. D Dorsal;
L left; R right; V ventral
100 µm
c
h
1 cm

100 µm
Cormidium

Female
Male i
Palpon
Bract
Gastrozooid
500 µm

A A
R L V D
P P

phylogeny of Dunn et al. (2005), was prepared for Forskalia formosa

SEM. The siphosomal growth zone was extremely dense,
with many tightly packed cormidia. There was a pro- SEM of the siphosomal growth zone of F. formosa indicates
nounced horn, and the young cormidia close to the tip that the zooids arise by probud subdivision (not shown),
were regular in organization and appeared to arise by although we did not describe the exact budding sequence.
probud subdivision (data not shown). The budding The mature cormidia consist of a posterior gastrozooid, a
sequence and mature organization of cormidia were not palpon, and a gonodendron bearing several gonopalpons
determined. and gonophores (Fig. 6). This is consistent with previous
Dev Genes Evol (2006) 216:743–754 751

A Discussion
D V Gonodendron
P Siphosomal elements arise independently
in the Cystonectae
Gonophore buds
Gastrozooids and gonodendra arise independently on the
Gonopalpon stem in the three long-stemmed cystonects examined here.
The arrangement of these elements can be either uniserial
Palpon or triserial (Fig. 2). The independent origin of the
gonodendra and gastrozooids, as well as the variable
number of gonodendra between gastrozooids within a
Bract specimen, bring into question the existence of cormidia in
a biologically relevant organizational sense in these taxa.
1 mm Gastrozooid There are five valid species of cystonects in total, two of
which were not examined. Of these, preliminary observa-
tions indicate that the siphosomal elements of Bathyphysa
Fig. 6 Schematic of a single cormidium of Forskalia formosa shown
from the right side. Multiple bracts are attached along the gastrozooid
conifera may arise in a uniserial pattern (P.R. Pugh,
peduncle (not shown). The gonodendron is immature. A Anterior; personal communication). Totton (1960) made an excellent
D dorsal; L left; P posterior; R right; V ventral description of the gross organization of Physalia physalis
(the Portuguese Man o’ War), the only cystonect that lacks
a long stem. Although his description of colony-level
descriptions of the cormidia of other species of Forskalia development is incomplete, it is sufficient to determine that
(Pugh 2003). The gastrozooids bear multiple bracts along budding in this species is substantially different from that in
their peduncles, and there are several small buds and the long-stemmed cystonects examined here.
juvenile bracts adjacent to the gastrozooid. The peduncle of
the palpon does not significantly elongate. The gonodendra Probud subdivision is a derived, shared mode
we observed were immature, so it was not possible to of development for the Codonophora
describe their organization in detail. It was clear that the
gonodendron arises from the base of the palpon. The The present study has described the origin of cormidia by
peduncle of the gonodendron appears to be an evagination probud subdivision in A. elegans and N. bijuga, and
of the stem, which at first bears only a single gonopalpon. established that it is the mode of cormidial development
Gonophores and further gonopalpons are added later in in F. formosa and an undescribed species of Apolemia. The
development. phylogenetic distribution of these taxa, and of those for
SEM of the nectosomal growth zone revealed a simple which previous descriptions of colony-level development
bud on the posterior base of the peduncle of each young are available, indicates that probud subdivision is a
nectophore, as was the case for A. elegans. synapomorphy of the Codonophora (Fig. 7).
Small buds were found to the posterior of young
Lychnagalma utricularia nectophores in both of the Codonophora taxa whose
nectosomal growth zones were examined (A. elegans and
We were able to make preliminary notes on the F. formosa). Such a bud was previously found in the
organization of the cormidia of L. utricularia, although nectosome of B. elongata (Dunn 2005), and the necto-
we have not yet looked at the siphosomal growth zone. We phores of Apolemia are known to alternate with nectosomal
found that the gastrozooids are borne on long peduncles, as polyps (Totton 1965). These findings indicate that the
had been noted previously (Pugh and Harbison 1986). We organization of the nectosome is more complex than had
also found that there is a palpon at the distal end of this previously been appreciated. It may be that the nectosome
peduncle just proximal to the gastrozooid. This gastric arose by tandem duplication of the siphosome and
palpon is associated with at least one bract, and several subsequent simplification.
other bracts are scattered along the length of the peduncle.
Palpons, male and female gonodendra, and bracts are The developmental underpinnings of organizational
attached to the stem between the gastrozooid peduncles. diversity in the Codonophora
There were no palpons within gonodendra of either sex.
Gastrozooids alternate with large palpons close to the The molecular phylogeny (Dunn et al. 2005) suggests a
growth zone. single transition from dioecy to monoecy within the
752 Dev Genes Evol (2006) 216:743–754

Cystonectae Codonophora
“Physonectae” Calycophorae
Gastrozooid bud
Pro-bud Pro-bud
Gonodendron bud
Each cormidium Each cormidium
Gastrozooids and arises as a single arises as a single
gonodendra arise pro-bud pro-bud
as seperate buds

Gonodendra buds The zooids of each The zooids of each

subdivide into cormidium arise by cormidium arise by
reproductive zooids subdivision of the subdivision of the
pro-bud pro-bud

Gonodendron The zooids of The zooids of

each cormidium each cormidium
Gastrozooid spread out along remain in a
the stem cluster

A
Cormidium
V D Cormidium
P

Chelophyes appendiculata
Stephanomia amphytridis

Sulculeolaria quadrivalvis
Physophora hydrostatica

Chuniphyes multidentata
Nectadamas diomedeae
Rhizophysa eysenhardti

Cordagalma cordiforme

Gymnopraia lapislazula

Sphaeronectes gracilis
Forskalia asymmetrica

Hippopodius hippopus
Athorybia rosacea Atl.
Bargmannia elongata

Nectopyramis natans
Bargmannia amoena
Rhizophysa filiformis
Bathyphysa sibogae

Abylopsis tetragona
Halistemma rubrum

Vogtia pentacantha

Clausophyes ovata

Muggiaea atlantica
Forskalia edwardsi
Forskalia tholoides

Forskalia formosa

Craseoa lathetica

Clausophyid sp 1
Physalia physalis

Rosacea flaccida

Lensia conoidea
Agalma elegans
Nanomia bijuga
Stephalia dilata

Diphyes dispar
Apolemia sp 1
Apolemia sp 4
Apolemia sp 2
Apolemia sp 3

Agalma clausi

Agalma okeni

Vogtia glabra
Praya dubia
Erenna sp

Phylogeny Key
Newly examined taxa
Previously examined taxa
Bayesian posterior > 0.90

Transition from Reiteration of Cormidial

Origin of dioecy to anterior-most paedomorphy;
Origin of nectosome,
siphosome, monoecy cormidial element Loss of
pro-bud pneumatophore
pneumatophore subdivision
Fig. 7 Hypothesized phylogenetic reconstruction of the history of basal polytomy in the sister group to Apolemia has been resolved to
several important transitions in siphonophore colony-level organi- show the monoecious species as being monophyletic. The previous
zation. Those taxa that have been examined in sufficient detail to molecular data were consistent with this topological hypothesis, but
determine the presence or absence of probud subdivision are did not favor it more than other hypotheses where monoecy arose
indicated with boxes. The rooted molecular phylogeny (using data more than once. Bathyphysa sibogae was not included in the
from 16S and 18S) is from Dunn et al. (2005). The topology of the molecular phylogenetic analysis; in this figure, it is shown with the
tree presented here differs from that in Dunn et al. (2005) in that the other cystonects

Codonophora (Fig. 7), although the relevant nodes are not tionally the same and differ only in the sex of the
well supported, and this hypothesis should be treated gonophores (Dunn 2005). Monoecious colonies are more
provisionally until more data are available. Within a complex in that both male and female gonophores must be
dioecious species, colonies of different sexes are organiza- organized along the stem. In the Calycophorae, each
Dev Genes Evol (2006) 216:743–754 753

cormidium bears only male or female gonophores (Totton Conclusions

1965). Within the monoecious “physonects”, each cormi-
dium bears male and female gonophores. The organiza- Comparisons between the Codonophora and Cystonectae,
tion and developmental origins of the male and female the broadest contrast that can be made among extant
elements are very different in the two monoecious siphonophores, reveal some key differences in colony-level
“physonects” examined in detail here, so better descrip- organization and development. The same two reiterated
tions of colony-level organization in other monoecious elements, the gastrozooids and gonodendra, are present in
“physonects” will need to be made before it is possible all long-stemmed cystonects. These elements arise inde-
to make specific hypotheses regarding the origin (or pendently, and there are a variable number of gonodendra
origins) of monoecy. between gastrozooids. The gonodendra, which consist of
The events following probud subdivision differ some- three types of zooids, are the most complex structures in the
what between the “Physonectae” and the Calycophorae. colony. The gonodendra arise by subdivision of the
In the “Physonectae”, the zooids within a cormidium gonodendron bud, but this subdivision has nothing to do
spread out along the stem, whereas in the Calycophorae, with the organizational context in which the gastrozooids
the zooids of a cormidium remain attached to the stem arise. In the Codonophora, the organizational context of all
by a common peduncle. The molecular phylogeny siphosomal zooids is established by the process of probud
indicates that the calycophoran state is derived (Fig. 7). subdivision that gives rise to cormidia. The cormidial
This derived calycophoran arrangement could have arisen organization of the Codonophora is extremely diverse
by cormidial paedomorphosis, whereby cormidial devel- across species (with respect to both the types of zooids
opment is truncated before the expansion of zooids along present and the position of the zooids relative to each
the stem, although the zooids within the cormidia mature other), although it is highly consistent within species.
fully. There is little known ecological diversity within the
Siphosomal budding has now been described in three long-stemmed Cystonectae (all species are piscivorous),
physonect species, B. elongata, A. elegans, and N. bijuga. and there are only five extant cystonect species in total
Agalma elegans and N. bijuga both belong to the (P. physalis, the fifth species, being clearly morphologically
Agalmatidae sensu stricto and are more closely related to distinct but still poorly known). The great morphological
each other than to B. elongata (Dunn et al. 2005). The diversity of the Codonophora is accompanied by large
cormidial organizations of A. elegans (Fig. 4c) and N. ecological diversity (with respect to prey and habitat), and
bijuga (Fig. 5c) are quite different. They are, however, the Codonophora are much more speciose than their sister
developmentally similar in that some zooids are intercalated group, the Cystonectae. These patterns in diversity make it
at well-defined locations within the cormidium and arise tempting to speculate that probud subdivision may have
directly on the stem. In both of these species, the been a key colony-level developmental innovation that
intercalated structures are reiterations of the anteriormost opened up a complex morphospace that is inaccessible to
derivative of the probud (this is the cluster consisting of a other colonial animals. Among other things, this morpho-
palpon, bracts, and male gonophores in A. elegans and the space includes an increased number of functionally spe-
cluster consisting of a palpon, male and female gonophores, cialized zooids and greater flexibility in the way these
and bracts in N. bijuga). Even when zooids arise directly on zooids are organized relative to each other.
the stem outside of the growth zone, they do so at well- It is possible that the organizational mechanisms at play
defined locations within the cormidia. in probud subdivision did not arise de novo but were
co-opted from zooid development. The development of a
Directional asymmetry of siphonophore colonies zooid from a bud (whether that bud arises independently
or by probud subdivision) requires the specification of
Dunn (2005) described the directional asymmetry of the fields that specify the location of various zooid structures.
cormidia of B. elongata, noting that many other descrip- It may be that these already-existing fields are used in the
tions of directional asymmetry are scattered throughout the probud to specify where the different zooids will arise
siphonophore systematics literature. No directional asymmet- during subdivision. Future mechanistic and molecular
ries were identified in any of the Cystonectae examined here. studies of colony-level and zooid-level development in
Within the Codonophora, no directional asymmetries were the siphonophores will provide an opportunity to test this
found in N. bijuga, but there was a bract on the left side of hypothesis.
each gastrozooid in A. elegans. Although data on directional
Acknowledgements This work was supported by a mini-PEET
asymmetries are still sparse, these findings suggest that
grant from the Society for Systematic Biology, a National Science
directional asymmetry has been gained and/or lost multiple Foundation Graduate Research Fellowship, and a National Science
times within the siphonophores. Foundation (NSF) doctoral dissertation improvement grant (DEB-
754 Dev Genes Evol (2006) 216:743–754

0408014). Thanks to S.H.D. Haddock, J. Godfrey, L. Madin, C. Hamner WM (1975) Underwater observations of blue-water plankton:
Sardet, and C. Carré for their assistance with collecting siphono- logistics, techniques, and safety procedures for divers at sea.
phores. E. Edwards provided valuable comments on the text. Special Limnol Oceanogr 20:1045–1051
thanks to P. Pugh for making his specimens available for this work, for Kawamura T (1910) “Bozunira” and “Katsuwo no Eboshi” Rhizophysa
providing feedback on the manuscript, for checking some of the and Physalia. Dobutz Z Tokyo 22:445–454
observations presented here, and, in particular, for our many Kawamura T (1911) “Shidarezakura Kurage” and “Nagayoraku
illuminating conversations. Kurage” Cupulita picta Metschnikoff and Agalmopsis elegans
Sars. Dobutz Z Tokyo 23:359–363
Leloup E (1936) Siphonophores calycophorides (suite) et physophor-
ides provenant des campagnes du Prince Albert Ier de Monaco.
Res Camp Sci Monaco 93:3–36
References
Maynard Smith J, Szathmáry E (1995) Major transitions in evolution.
Oxford University Press, Oxford
Beklemishev WN (1969) Principles of comparative anatomy, vol 1. Pugh PR (1984) The diel migrations and distributions within a
Promorphology. Oliver and Boyd, Edinburgh mesopelagic community in the north east Atlantic. 7. Siphono-
Bonner J (2001) First signals: the evolution of multicellular phores. Prog Oceanogr 13:461–489
development. Princeton University Press, Princeton Pugh PR (2003) A revision of the family Forskaliidae (Siphonophora,
Buss L (1987) The evolution of individuality. Princeton University “Physonectae”). J Nat Hist 37:1281–1327
Press, Princeton Pugh PR, Harbison GR (1986) New observations on a rare physonect
Carré C, Carré D (1993) Ordre des Siphonophores. In: Doumenc D siphonophore, Lychnagalma utricularia (Claus, 1879). J Mar
(ed) Traité de Zoologies: Anatomie, Systematique, Biologie, Biol Assoc UK 66:695–710
vol 3(2). Masson, Paris, pp 523–596 Robison BH (1995) Light in the ocean’s midwaters. Sci Am 273:
Chun C (1885) Über die cyklische Entwickelung der Siphonophoren. 60–64
Sitz Ber K Preuss Akad Wiss Berl 1885:511–529 Schneider KC (1896) Mittheilungen über Siphonophoren. II. Grundriss
Dunn CW (2005) The complex colony-level organization of the deep- der organization der Siphonophoren. Zool Jahrb Abt Anat Ontog
sea siphonophore Bargmannia elongata (Cnidaria, Hydrozoa) is Tiere 9:571–664
directionally asymmetric and arises by the subdivision of pro- Totton AK (1954) Siphonophora of the Indian Ocean. Discov Rep
buds. Dev Dyn 234:835–845 27:1–162
Dunn CW, Pugh PR, Haddock SHD (2005) Molecular phylogenetics Totton AK (1956) Development and metamorphosis of the larva of
of the Siphonophora (Cnidaria), with implications for the Agalma elegans (Sars) (Siphonophora “Physonectae”). Pap Mar
evolution of functional specialization. Syst Biol 54:916–935 Biol Oceanogr 3(Suppl):239–241
Gegenbaur C (1853) Beiträge zur näheren Kenntniss der Schwimm- Totton AK (1960) Studies on Physalia physalis. Part 1. Natural history
polypen (Siphonophoren). Arch Naturgesch 5:285–344 and morphology. Discov Rep 30:301–368
Haddock SHD, Dunn CW, Pugh PR (2005) A reexamination of Totton AK (1965) A synopsis of the siphonophora. British Museum of
siphonophore terminology and morphology, applied to the Natural History, London
description of two new prayine species with remarkable bio- Youngbluth M (1984) Manned submersibles and sophisticated instru-
optical properties. J Mar Biol Assoc UK 85:695–707 mentation: tools for oceanographical research. In: “Proceedings of
Haeckel E (1869) Zur Entwickelungsgeschichte der Siphonophoren. SUBTECH ’83 Symposium.” Society for Underwater Technology,
Natuurk Verh Prov Utrechtsh Genoots 6:1–120 London, pp 335–344