Commemorative volume for the 80th birthday of Willem Vervoort in 1997

Heterochrony, generic distinction and phylogeny in the family Hydractiniidae (Hydrozoa: Cnidaria)

F. Boero, J. Bouillon & S. Piraino

Boero, F., J. Bouillon & S. Piraino. Heterochrony, generic distinction and phylogeny in the family Hydractiniidae (Hydrozoa: Cnidaria). Zool. Verh. Leiden 323, 31.xii.1998: 25-36, figs 1-4.— ISSN 0024-1652/ISBN 90-73239-68-0. F. Boero, Dipartimento di Biologia, Stazione di Biologia Marina, Università di Lecce, 73100 Lecce, Italy. J. Bouillon, Laboratoire de Biologie Marine, Université Libre de Bruxelles, 50 Ave F.D. Roosevelt, 1050 Bruxelles, Belgium. S. Piraino, Istituto Sperimentale Talassografico del CNR, Via Roma 3, 74100 Taranto, Italy.

Key words: Cnidaria; Hydrozoa; Anthomedusae; phylogeny; life cycle; classification; heterochrony; paedomorphosis. The taxonomy of Hydractinia, Stylactaria and Podocoryna is discussed and the three genera are merged into Hydractinia since their diagnostic characters are liable to lead to polyphyly and paraphyly, due to repeated episodes of medusa reduction via heterochrony (paedomorphosis). The phylogeny of the Hydractiniidae is reconstructed by using two outgroups, Clava and Cytaeis, both having some charac­ters in common with the Hydractiniidae. The resulting phylogenetic trees agree in identifying affini­ties among Hydractinia, Kinetocodium and Hydrocorella, all with polymorphic colonies with gastro-zooids having oral tentacles. The position of Clavactinia (characterized by gastrozooids with widely scattered tentacles) is at the root of the tree if Clava is the outgroup, whereas it becomes apical when the outgroup is Cytaeis. The pattern of medusa suppression is different in the two cladograms, since the presence of a medusa is a plesiomorphic feature when Cytaeis is the outgroup, whereas it becomes apomorphic when the outgroup is Clava. These inconveniences are difficult to accommodate, since medusa suppression has occurred many times in the evolution of the hydroidomedusae, and Recent species do not witness past paedomorphic events of medusa reduction properly, so that many inter­mediate states are probably missing.

Introduction

T w o genera of the f a m i l y H y d r a c t i n i i d a e are w i d e l y used i n experimental b io lo­

gy, namely Hydractinia a n d Podocoryna. A t h i r d a l l ied genus, m u c h less used for

experimental research, is Stylactaria.

Previous studies of species of these genera have concentrated o n regeneration (e.

g., A c h e r m a n n , 1980), embryology (e.g., B o d o & B o u i l l o n , 1968; K r o i h e r & Plickert ,

1992), myogenesis, gametogenesis a n d cnidogenesis (e.g., Boelsterli , 1977), i n d u c t i o n

of sexuality (e.g., Braverman, 1962), g r o w t h (e.g., Berr i l l , 1953; B r a v e r m a n & Schrandt,

1969), tumor p r o m o t i o n (e.g., K u r t z & S c h m i d , 1991), expression of homeobox genes

(e.g., A e r n e et al . , 1995, 1996), transdifferentiation (e.g., S c h m i d , 1988; S c h m i d et a l ,

1988; Reber-Müller et a l , 1994), the organizat ion of the extracellular matr ix (e.g.,

Weber & S c h m i d , 1985; Reber-Müller et al . , 1996), the c o u p l i n g of ontogeny a n d p h y ­

logeny (e.g., Blackstone & Buss, 1992), the c o u p l i n g of geological a n d molecular data

i n the reconstruction of symbiot ic relationships (e.g., C u n n i n g h a m et al . , 1991), and

paedomorphosis (e.g., C u n n i n g h a m & Buss, 1993; Blackstone & Buss, 1993).

In spite of these contributions to experimental b io logy, a n d of be ing a classical

26 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

example of p o l y m o r p h i s m i n most textbooks o n C n i d a r i a , the H y d r a c t i n i i d a e st i l l do not have a settled generic taxonomy. C u n n i n g h a m & Buss (1993) inferred relation­ships f r o m molecular data of species of Hydractinia, Podocoryna and Stylactaria to test the phylogenetic m e a n i n g of medusa reduct ion i n a l l hydro idomedusae . They s h o w e d that the three genera of the fami ly are probably not monophylet ic , an o p i n i o n shared also b y Boero et al . (1996) f r o m a p r e l i m i n a r y comparative analysis of m o r p h o ­logical characters. These f indings support Petersen's (1990) cr i t ic ism of the use of medusa suppress ion as a generic character.

Species, furthermore, are not easily ident i f ied a n d this is the first group of h y d r o -zoans i n w h i c h s ib l ing species have been recognized (Buss a n d Y u n d , 1989), so that m a n y experimental studies might have been made o n material of dubious identi ty.

The characters used to d is t inguish Hydractinia, Podocoryna and Stylactaria w i l l be r e v i e w e d to attempt a formal isat ion of shortcomings of current taxonomy, and to suggest evolut ionary hypotheses that w i l l be testable i n future w o r k w i t h the a i m of constructing a phylogenetic classification.

The morphological and phylogenetic basis for generic boundaries in Hydractinia, Podocoryna and Stylactaria

The genera Hydractinia, Podocoryna and Stylactaria are d is t inguished b y the states of t w o m a i n morpholog ica l characters: 1 - the structure of the h y d r o r h i z a i n f u l l y g r o w n colonies (fig. 1): • reticular, formed b y anastomosed stolons w h i c h remain separated (Stylactaria and

some Podocoryna), • encrusting, f o r m e d b y anastomosed stolons covered b y a c o m m o n perisarc (some

Podocoryna), • encrusting, f o r m e d b y anastomosed stolons covered b y naked coenosarc after

degeneration of the upper part of the stolonal perisarc sheath (Hydractinia and some species of Podocoryna).

2 - the structure of the gonosome: • free medusae (Podocoryna). • f ixed or libérable eumedusoids (some species of Hydractinia a n d Stylactaria). • f ixed sporosacs (some species Hydractinia a n d Stylactaria).

The structure of the h y d r o r h i z a , reticular or encrusting, has been generally accept­ed b y recent authors as a generic character i n the H y d r a c t i n o i d e a ( M i l l a r d , 1975; B o u i l l o n , 1985; 1995; Calder , 1988; H i r o h i t o , 1988; Schuchert, 1996). This character, however , seems to depend sometimes o n the nature of the substrate (see E d w a r d s , 1972; Jarms, 1987; H i r o h i t o , 1988). In m a n y y o u n g or regenerating colonies of Hydrac­tinia the hydrorh izae are reticulate but, nevertheless, the potential for the p r o d u c t i o n of a layer of naked coenosarc is retained. For Blackstone & Buss (1991; 1992) reticulate forms are paedomorphic (progenetic) compared w i t h encrust ing ones. The g r o w t h of a h y d r a c t i n i i d colony, i n fact, a lways starts w i t h a reticulate h y d r o r h i z a (thus, a juve­nile character) w h i c h then becomes encrusting b y anastomosing.

A c c o r d i n g to 'classic' views o n hydroidomedusan phylogenetic characters (reviewed b y C u n n i n g h a m & Buss, 1993), the presence of free medusae is a p les iomorphic fea­ture, whereas medusa reduct ion represents a der ived state. In this f ramework, Podoco-

Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998) 27

Podocoryna - Stylactaria non-encrusting, reticular hydrorhiza formed by anastomosing stolonal tubes surrounded by perisarc

Podocoryna encrusting hydrorhiza formed by coalescent stolons leading to a perisarc-covered mat

Podocoryna - Hydractinia encrusting hydrorhiza formed by coalescent stolons whose perisarc degenerates in the upper portion leading to a coenosarc-covered mat

Fig. 1. Schematic representation (in section) of hydrorhizal features of the three main genera of the Hydractiniidae. Black: perisarc; dark grey: surface of transverse hydrorhizal tube connecting two main hydrorhizal tubes; light grey: coenosarc.

ryna s h o u l d have p les iomorphic features l i n k e d to the presence of free medusae and, also, to the possession of h y d r o r h i z a l character states cover ing a l l the possibil i t ies list­ed above, be ing either reticular, perisarc- or coenosarc-covered mat- l ike. Stylactaria s h o u l d represent a monophyle t i c clade w i t h medusae reduced to eumedusoids , and a reticular h y d r o r h i z a (both features as a result of paedomorphosis) . Hydractinia s h o u l d be a monophyle t i c clade w i t h medusae reduced to paedomorphic eumedusoids or to sporosacs, a n d w i t h stolonal differentiation into an elaborated mat of coenosarc, a p les iomorphic feature present also i n some Podocoryna.

The morpholog ica l overlaps i n generic characters a m o n g the three taxa, a long w i t h the incongruences stressed b y C u n n i n g h a m & Buss (1993) u s i n g molecular data, cal l for reconsideration of the value g i v e n to generic characters i n reconstructing p h y -logenies.

Sibling species and sibling genera, or: are ancestors real?

A s s i g n m e n t of species to separate genera o n the basis of d i f fer ing medusa expres­s ion (due to paedomorphosis) has been crit icised b y Petersen (1979; 1990), since pae­d o m o r p h i c patterns have been s h o w n i n a few genera to have arisen more than once. This v i e w has been supported w i t h i n the H y d r a c t i n i i d a e b y molecular evidence pro­v i d e d b y C u n n i n g h a m & Buss (1993). The same argument m i g h t be used for the type

28 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

of h y d r o r h i z a l organisation, so that colonies w i t h paedomorphic features (such as those assigned to Stylactaria) might be the result of independent phenomena of hete­rochrony affecting the expression of bo th medusa and h y d r o r h i z a .

It might be possible that a paedomorphic species d e r i v i n g f r o m a Podocoryna-like ancestor originated a monophyle t i c clade w h i c h c o u l d be recognised as a s o u n d genus. But paedomorphic species d e r i v i n g f r o m other Podocoryna-like ancestors might evolve the same m o r p h o l o g y independent ly , so that one c o u l d consider them as be longing to ' s ib l ing genera' (fig. 2).

U n d e r such circumstances, the monophyle t i c clade might st i l l be considered as a genus, but the isolated paedomorphs s h o u l d remain connected to the 'ancestral ' genus. Petersen (1979, 1990) argued that i n a few genera medusa loss has probably occurred more than once w i t h i n the same clade, thus leading to po lyphyle t i c taxa as based o n a generic character not correlated w i t h other autapomorphies . For this rea­son the paedomorphs are to be assigned to the same genus of the non-paedomorphic species, if no other characters diverge f r o m the ancestral state. H o w e v e r , if, as sug­gested above, an event of medusa reduct ion were f o l l o w e d b y radiat ion of species retaining that character, the result might be a monophyle t i c clade of possible generic rank.

In a l l evolut ionary studies based o n l i v i n g species, it is almost impossible to recognise sister species f r o m 'mother ' species. C l a d i s m , furthermore, considers ances­tors as ideal ised entities w i t h p les iomorphic features, whereas it is undeniable that ancestors are not ideas but species, at least as m u c h as their descendants (Rasnitsyn, 1996). It is not inconceivable that a st i l l l i v i n g species originated a gro up of species b y cladogenesis, w h i l e remain ing u n m o d i f i e d or go ing through anagenesis i n its core population(s) . These events, i n both morpho lo g ic a l a n d molecular studies, might lead to identi f icat ion of ' s is terhood' and not of 'motherhood ' . Palaeontology might help i n s o l v i n g the problem, but the fossil record of hydract in i ids is too scarce to be of use.

The lack of consistency f o u n d b y C u n n i n g h a m & Buss (1993) w h i l e c o m p a r i n g genome fractions of Hydractinia, Podocoryna and Stylactaria might suggest that single species assigned either to Stylactaria a n d Hydractinia or iginated independent ly f r o m Podocoryna-like ancestors, l eading to po lyphyle t i c {Stylactaria a n d Hydractinia) a n d paraphylet ic (Podocoryna) taxa, as suggested b y Boero et al . (1996) (fig. 2).

Be ing 'p les iomorphic ' , the Podocoryna set of characters might lead to g r o u p i n g species into a compact clade, also f r o m a molecular point of v i e w . The compactness of the clade, however , s h o u l d not be necessarily an indica t ion of m o n o p h y l y , but on ly of lack of devia t ion f r o m an ancestral state.

The c ladogram proposed i n f ig . 2 is rather unorthodox, since it represents ances­tors as real taxa. Furthermore, it is characterised b y several paral le l events of e v o l u ­t ionary change. H o w e v e r , if a change is due to a d d i t i o n of n e w characters (e.g., the acquis i t ion of a medusa stage), it is unpars imonious to propose independent origins of the same character, but if the change is due character s impl i f i ca t ion (e.g., the loss of a medusa stage v i a paedomorphosis) , the possibi l i ty of paral le l events is not to be considered as unpars imonious as i n the former case. M e d u s a suppress ion has occurred i n the H y d r o i d o m e d u s a e independent ly so m a n y times (see B o u i l l o n , 1985; Petersen, 1990; Cornel ius , 1992) that is currently considered as a very probable event.

Current ly , the o n l y w a y to a v o i d the unnatura l taxa d e r i v i n g f r o m the s i tuation

Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998) 29

Fig. 2. Hypothetical phylogenetic scenario showing the possibility of the polyphyly of the currently recognised hydractiniid genera with reduced medusae (Stylactaria and Hydractinia), and of the para-phyly of the genus comprising species with medusae (Podocoryne).

depicted i n f ig . 2 is to merge both polyphyle t i c a n d paraphylet ic genera, so as to have a monophyle t i c clade, even if of very remote ancestry. So w e propose here to merge the m a i n three genera of hydrac t in i ids (Hydractinia v a n Beneden, 1841; Podocoryna Sars, 1846; Stylactaria Stechow, 1921) into the oldest one: Hydractinia v a n Beneden, 1841. S imi lar posit ions have been taken already b y several authors: M o t z - K o s s o w s k a (1905) merged Hydractinia, Podocoryna and Stylactaria (= Stylactis) into Hydractinia;

30 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

Broch (1914) considered that Stylactaria a n d Hydractinia c o u l d not be kept apart; and N a u m o v (1960/1969) merged Podocoryna into Hydractinia. K r a m p (1932) gave Podoco­ryna a n d Stylactaria subgeneric rank w i t h i n Hydractinia, but this is not phylogenet ical -l y sound. If Podocoryna and Stylactaria were monophylet ic , they c o u l d w e l l be ranked as genera; if they are not monophyle t i c (as it is suggested here), they are u n s o u n d even as subgenera.

These inconveniences cannot be a v o i d e d i n morpholog ica l studies; molecular approaches might he lp i n dist inct ion of monophyle t i c clades f r o m 'sister clades' shar­i n g medusa reduct ion, thus ident i fy ing what c o u l d be possibly cal led ' s i b l i n g genera'.

T h e d iagnos is of Hydractinia v a n Beneden , 1841

Colonies w i t h a stolonal reticular h y d r o r h i z a f o r m e d b y tubes covered w i t h peris­arc, or w i t h an encrusting h y d r o r h i z a covered w i t h perisarc or w i t h naked coenosarc; typica l ly w i t h s imple, canaliculated or branched spines. H y d r a n t h s sessile, naked, p o l y m o r p h i c : gastrozooids, gonozooids , and occasionally dactylozooids . Gastro-zooids w i t h one or more close w h o r l s of tentacles encirc l ing the hypostome; gono­zooids w i t h one or more close w h o r l s of tentacles, or w i t h o u t tentacles and/or h y p o ­stome, be ing reduced to blastostyles; dactylozooids w i t h o u t tentacles. Gonophores : either f ixed sporosacs, libérable or retained eumedusoids , or free medusae, ar is ing f r o m v a r y i n g l y developed gonozooids or direct ly f r o m the h y d r o r h i z a . M e d u s a e more or less bel l -shaped. Four radia l canals; four or more so l id and s imple m arg ina l tentacles, w i t h or w i t h o u t ocell i . M a n u b r i u m w i t h or w i t h o u t peduncle , tubular or sac-shaped, w i t h four or eight s imple or s l ight ly branched oral arms (dilatations of the m o u t h r im) armed w i t h clusters of cnidocysts. Gonads o n m a n u b r i u m , generally interradial but sometimes extending to the p r o x i m a l por t ion of radia l canals. Some­times asexual reproduct ion b y medusa b u d d i n g o n m a n u b r i u m .

T h e p h y l o g e n y of the H y d r a c t i n i i d a e

B o u i l l o n (1995) assigned the f o l l o w i n g l i v i n g genera to the H y d r a c t i n i i d a e : Clavac-tinia Thornely , 1904; Hansiella B o u i l l o n , 1980; Hydractinia v a n Beneden, 1841; Hydro-corella Stechow, 1921; Janaria Stechow, 1921; Kinetocodium K r a m p , 1921; Podocoryna Sars, 1846; Stylactaria Stechow, 1921; and Tregoubovia P i card , 1958. B o u i l l o n (1985; 1995) also l isted a series of fossil genera do ubt fu l ly assigned to this fami ly because of skeletal remains, but the pauci ty of diagnostic features makes them useless for a p h y ­logenetic reconstruction of the f a m i l y a n d they w i l l not be considered here.

Both Hansiella and Tregoubovia are medusa-based genera, their h y d r o i d s be ing u n k n o w n . The medusa of Thecocodium quadratum (Werner, 1965) (family P t i locodi -idae) described b y Jarms (1987), however , shares some key features w i t h the medusae of the t w o above ment ioned genera, namely a m arg ina l nematocyst r i n g a n d dider-m i c exumbrel lar processes, and this w i l l a l l o w assignment of both Hansiella a n d Tre­goubovia to the Pt i locodi idae (Boui l lon et al . , 1997). Hydractinia, Podocoryna a n d Stylac­taria are here merged into Hydractinia (see above). Hydrocorella and Janaria are charac­terized b y a calcareous skeleton. Generic differences are m i n o r for the purpose of this revis ion a n d the t w o genera w i l l be treated together under Hydrocorella even if they

Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998) 31

are to be considered as separate (see Cairns & Barnard, 1984 for a treatment of Janaria). Kinetocodium comprises o n l y one species w h i c h has been f o u n d a few times, epizoic o n pteropods ( K r a m p , 1921; 1957). O n l y the y o u n g m e d u s a is k n o w n a n d its pos i t ion is s t i l l uncertain, p e n d i n g k n o w l e d g e of its complete life cycle. Thus , for ease of analysis, the H y d r a c t i n i i d a e is considered here to comprise Clavactinia, Hydractinia, Kinetocodium and Hydrocorella.

The outgroups of the H y d r a c t i n i i d a e

The key features of the H y d r a c t i n i i d a e (see above for details) are the presence of naked hydranths d e p r i v e d of stem, direct ly connected to a h y d r o r h i z a l system w h i c h can be a more or less elaborated network. The hydranths are p o l y m o r p h i c , be ing dis­t inguished i n tentacled gastrozooids, and gonozooids . In m a n y species both dactylo­zooids a n d spines are present. The medusae, w h e n present, have oral l ips a rmed w i t h nematocysts w h i c h can be arranged i n clusters (Podocoryna) or i n a single continuous r o w (Kinetocodium).

A survey of a l l other A n t h o m e d u s a e ( from B o u i l l o n , 1985) for the choice of an outgroup indicates t w o genera as best candidates: • Clava ( family Clavidae) shares w i t h the H y d r a c t i n i i d a e the presence of naked

hydranths d e p r i v e d of stem, direct ly connected to a h y d r o r h i z a l system w h i c h can range f r o m s imple stolons to a ne twork of t ight ly anastomosed stolons some­times becoming covered, as reported b y H i n c k s (1868) a n d M o t z - K o s s o w s k a (1905), b y a layer of perisarc. Contrary to the H y d r a c t i n i i d a e , Clava is m o n o m o r ­phe a n d its gonophores, as f ixed sporosacs, originate o n undifferentiated h y ­dranths. Fraser (1946: 138) considered that Stylactaria "must have come directly f r o m the Clav idae" .

• Cytaeis ( family Cytaeidae) shares w i t h the H y d r a c t i n i i d a e the presence of naked hydranths d e p r i v e d of stem, direct ly connected to a reticular h y d r o r h i z a , but is m o n o m o r p h i c , a n d the gonophores arise f r o m the h y d r o r h i z a . The medusae have oral tentacles inserted above the m o u t h .

The p o l a r i s a t i o n of the H y d r a c t i n i i d a e

Clava as an o u t g r o u p . — The h y d r a c t i n i i d genus w i t h characters closest to Clava is Clavactinia. In both genera the hydranths are naked a n d have scattered tentacles, but Clavactinia is p o l y m o r p h i c and its h y d r o r h i z a is covered b y naked coenosarc (see M i l ­l a r d & B o u i l l o n , 1973 for a detailed description). Both genera lack a medusa stage. A l l the other genera of the H y d r a c t i n i i d a e have tentacles concentrated a r o u n d the m o u t h , i n one or m u l t i p l e but closely set w h o r l s (for this reason w e inc lude Clavactinia multi-tentaculata M i l l a r d , 1975, i n Hydractinia). Hydractinia, w i d e l y discussed above, is char­acterised b y p o l y m o r p h i s m , gastrozooids w i t h ora l tentacles, a n d b y the presence of medusae w i t h oral l ips a rmed w i t h nematocyst clusters. Kinetocodium is kept separate f r o m Hydractinia because gonozooids are absent, a n d its medusae originate directly f r o m the h y d r o r h i z a and have a continuous r i n g of nematocysts a r o u n d the m o u t h , instead of nematocyst clusters. Hydrocorella has p o l y m o r p h i c colonies w i t h f ixed gonophores a n d is separated because its h y d r o r h i z a is calcareous, resembling that of

32 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

Fig. 3. Cladogram of the Hydractiniidae with Clava as an outgroup.

a h a r d coral . The result ing c ladogram is presented i n f ig . 3. Cytaeis as an o u t g r o u p . — Both Cytaeis a n d some species of Hydractinia (those for­

mer ly assigned to Podocoryna) share the presence of a medusa . The m a i n difference is that i n Cytaeis the oral zone of the m a n u b r i u m is armed w i t h ora l tentacles inserted above the m o u t h , whereas i n Hydractinia the m o u t h is armed w i t h more or less devel ­oped nematocyst clusters borne direct ly o n the l ips . Besides this feature, the medusae of the t w o families are quite s imilar . The h y d r o r h i z a of Cytaeis is reticular, as is that of some species of Hydractinia. The hydranths , furthermore, are s imilar , even though Cytaeis is m o n o m o r p h i c . A s i n the c ladogram result ing f r o m the choice of Clava as an outgroup (fig. 3), the genera Kinetocodium a n d Hydrocorella are close to Hydractinia, whereas Clavactinia is separated f r o m the rest of the clade b y the presence of scattered tentacles. The result ing c ladogram is g i v e n i n f ig . 4.

The compar ison of the c ladograms d e r i v i n g f r o m the t w o chosen outgroups (figs. 3 and 4) s h o w that w h e n Clava is the outgroup, Clavactinia has p les iomorphic fea­tures, whereas its scattered tentacles become apomorphic if Cytaeis is the outgroup. The presence of a m e d u s a stage, w h e n Clava is the outgroup, is not connected w i t h ancestral features, a n d this leads to the hypothesis that ancestral species w i t h medusae and w i d e l y scattered tentacles became extinct or are st i l l to be discovered. The hypothesis is supported b y the presence of t w o genera w i t h medusae (Turritopsis and Oceania) i n the Clav idae . Their h y d r o i d s have erect, branched stems, be ing noticeably different f r o m those of the H y d r a c t i n i i d a e , but their medusae are very s i m ­ilar to those of the H y d r a c t i n i i d a e , a n d there m i g h t be n o justif ication for keeping them i n separate families, if their h y d r o i d s were not so different.

Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998) 33

Fig. 4. Cladogram of the Hydractiniidae with Cytaeis as an outgroup.

W h e n Cytaeis is the outgroup, there is no need for the extinction of ancestors w i t h medusae, since both Cytaeis a n d some Hydractinia have medusae and naked feeding hydranths w i t h a single r o w of tentacles, or ig inat ing f r o m a reticular h y d r o r h i z a . The m a i n difference i n the medusae of Cytaeis a n d Hydractinia is the presence of oral ten­tacles i n the former. The passage f r o m Cytaeis to Hydractinia leads one to hypothesise the loss of the oral tentacles (reduced to nematocyst clusters) a n d the evo lut ion of p o l y m o r p h i s m .

Conclusion

The scope of Hydractinia s t i l l remains doubt fu l , due to lack of in format ion about the genera here merged w i t h it: Podocoryna a n d Stylactaria. It m i g h t be possible that monophyle t i c clades of species fa l l ing w i t h i n the scope of Hydractinia a n d Stylactaria real ly exist, but that repeated heterochronic events i n Podocoryna-like ancestors l ed to ' s ib l ing genera'. The H y d r a c t i n i i d a e have strict affinities w i t h both the C l a v i d a e and the Cytaeidae.

The p r o b l e m of m e d u s a loss raises some problems i n phylogenetic analysis. It is conceivable, that a l l families w i t h medusae originated f r o m ancestors w i t h medusae, a n d that medusae were lost independent ly i n the var ious families. The hypothesis that such elaborate m o r p h s as the medusae e v o l v e d independent ly i n m a n y clades w i t h no medusa requires far too u n l i k e l y paral le l events. Some families, furthermore, have no species w i t h medusae. M e d u s a reduct ion or loss b y paedomorphosis has occurred repeatedly a n d independent ly d u r i n g h y d r o i d o m e d u s a n evolut ion result ing

34 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

i n s imi lar i ty i n m o r p h o l o g y perhaps being unrel iable as a point of reference i n some genera. The g r o u p i n g of al l species w i t h reduced medusae i n clades separate f r o m those of s imi lar species w i t h medusae, i n fact, is more pars imonious than consider ing them as originated b y possible paral le l events, but this is the case i n a few k n o w n examples (as convinc ingly argued b y Petersen, 1990), i n spite of being less pars imo­

nious. C o m p a r i s o n of the t w o cladograms der ived u s i n g different outgroups (figs 3 and

4), i n this f ramework, makes Cytaeis a better outgroup than Clava, because i n this case both the outgroup and the stem species of the fami ly have the ples iomorphic feature of a medusa stage. W h e n Clava is the outgroup, the medusa appears as a n e w l y ­

acquired feature, a less pars imonious conclusion, as hypothes ized above (but medusae c o u l d become suppressed i n some clades w h i l e h a v i n g the possibi l i ty of being re­expressed i n the course of evolut ion, as argued b y Boero et al . , 1996). Clava, furthermore, is referred to the distant fami ly C l a v i d a e solely o n the basis of its scat­

tered tentacles, but such a feature is p r o v i n g to be shared b y m a n y species referred to different families and is possibly not a good f a m i l y character. The h y d r o r h i z a l system of Clava closely resembles that of some Hydractinia species a n d future molecular s tud­

ies w i l l possibly better define the affinities of these taxa a n d the w h o l e p h y l o g e n y of the Hydrac t in idae .

Acknowledgements

This s tudy was supported b y contributions f r o m M . U . R . S . T . (40 and 60% pro­

grams), the M a r i n e Reserve of Ust ica (Italy), Province of Lecce, a n d of the F . N . R . S . de Belgique. Leo Buss and Cl i f f C u n n i n g h a m read a n d commented o n the manuscript , sharing w i t h us their precious 'molecular ' insights.

This paper is dedicated to our f r iend and colleague W i m Vervoort , whose w o r k o n hydrozoans w i l l r e m a i n an invaluable explorat ion of the divers i ty of the group.

References

Achermann, J., 1980. The fate and regeneration capacity of isolated ecto­ and endoderm in polyps of Podocoryne carnea M . Sars (Hydrozoa, Athecata). In: P. Tardent & R. Tardent (eds). Developmental and cellular biology of coelenterates.— Elsevier/North­Holland Biomedical Press, Amsterdam: 273­279.

Aerne, B.L., C D . Baader & V. Schmid, 1995. Life stage and tissue­specific expression of the homeobox gene cnoxl­Pc of the hydrozoan Podocoryne carnea.— Dev. Biol. 169: 547­556.

Aerne, B., H . Gròger, P. Schuchert, J. Spring & V. Schmid, 1996. The polyp and its medusa: a molecu­

lar approach.— Sei. Mar. 60 (1): 7­16. Berrill, N.J. 1953. Growth and form in gymnoblastic hydroids. VI. Polymorphism within the Hydrac­

tiniidae.—J. Morph. 92: 241­272. Blackstone, N.W. & L. Buss, 1991. Shape variation in hydractiniid hydroids.— Biol. Bull. (Woods

Hole) 180 (3): 394­405. Blackstone, Ν. W. & L. Buss, 1992. Treatment with 2, 4­dinitrophenil mimics ontogenetic and phylo­

genetic changes in a hydractiniid hydroid.— Proc. Nat. Acad. Sei. U.S.A. 89: 4057­4061. Blackstone, Ν. W. & L. W. Buss, 1993. Experimental heterochrony in hydractiniid hydroids: why

mechanisms matter.— J. Evol. Biol. 6 (3): 307­327. Bodo, F. & J. Bouillon, 1968. Étude histologique du développement embryonnaire de quelques

Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998) 35

hydroméduses de Roseoff: Phialidium hemisphaericum (L.), Obelia sp. Péron et Lesueur, Sarsia eximia (Allman), Podocoryne carnea (Sars), Gonionemus vertens Agassiz.— Cah. Biol. Mar. 9: 69-104.

Boelsterli, U . 1977. A n electron microscopic study of early developmental stages, myogenesis, oogene­sis and cnidogenesis in the Anthomedusa, Podocoryne carnea M . Sars.— J. Morph. 154: 259-290.

Boero, F., J. Bouillon & S. Piraino, 1996. Classification and phylogeny in the Hydroidomedusae (Hydrozoa, Cnidaria).— Sei. Mar. 60 (1): 17-33.

Bouillon, J. 1985. Essai de classification des Hydropolypes - Hydroméduses (Hydrozoa-Cnidaria).— Indo-Malayan Zool. 2: 29-243.

Bouillon, J. 1995. Classe des Hydrozoaires. In: P.P. Grasse & D. Doumenc (eds). Traité de Zoologie 3.— Masson, Paris: 29-416.

Bouillon, J., D. Medel & A.L. Pena Cantero, 1997. The taxonomie status of the genus Stylactaria Ste­chow, 1921 (Hydroidomedusae, Anthomedusae, Hydractiniidae) with the description of a new species.— Sei. Mar. 61 (4): 471-486.

Braverman, M . H . 1962. Studies in hydroid differentiation. I. Podocoryne carnea culture methods and carbon dioxide induced sexuality.— Exp. Cell Res. 26: 301-306.

Braverman, M . H . & R. Schrandt, 1969. Studies on hydroid differentiation. V. The control of growth in young colonies of Podocoryne carnea.— Growth 33: 241-254.

Broch, H . 1914. Hydrozoa benthonica. In: W. Michaelsen (ed.). Beiträge zur Kenntnis der Meeresfauna westafrikas. 1.— Friederichsen, Hamburg: 19-50.

Buss, L. & P. Yund, 1989. A sibling species group of Hydractinia in the north-eastern United States.— J. Mar. Biol. Ass. U . K. 69: 857-874.

Cairns, S.D. & J.L. Barnard, 1984. Redescription of Janaria mirabilis, a calcified hydroid from the east­ern Pacific— Bull. S. Calif. Acad. Sei. 83:1-11.

Calder, D. 1988. Shallow water hydroids of Bermuda. The Athecatae.— Life Sei. Contr. R. Ontario Mus. 148:1-107.

Cornelius, P.F.S., 1992. Medusa loss in leptolid Hydrozoa (Cnidaria), hydroid rafting, and abbreviated life-cycles among their remote-island faunae: an interim review.— Sei. Mar. 56 (2-3): 245-261.

Cunningham, C , L.W. Buss & C. Anderson, 1991. Molecular and geologic evidence of shared history between hermit crabs and the symbiotic genus Hydractinia. — Evolution 45:1301-1315.

Cunningham, C. & L.W. Buss, 1993. Molecular evidence for multiple episodes of paedomorphosis in the family Hydractiniidae.— Biochem. Syst. Ecol. 21: 57-69.

Edwards, C. 1972. The hydroids and the medusae Podocoryne areolata, P. borealis and P. carnea.— J. Mar. Biol. Ass. U . K. 52: 97-144.

Fraser, C M . 1946. Distribution and relationship in American hydroids.— The Univ. Toronto Press, Toronto: 1-464.

Hincks, T. 1868. The History of the British hydroid zoophytes.— John Van Voorst, London, Vol. 1: 1-338, Vol. 2: 67 plates.

Hirohito, Emperor of Japan, 1988. The Hydroids of Sagami Bay. Part 1. Athecata.— Pubis Biol. Lab., Imp. Household, Tokyo: 1-179.

Jarms, G. 1987. Thecocodium quadratum (Werner, 1965) redescribed, T. penicillatum sp. nov., and a method for rearing hydrozoans. In: J. Bouillon, F. Boero, F. Cicogna, & P.F.S. Cornelius (eds). Modern trends in the Systematics, Ecology and Evolution of Hydroids and Hydromedusae.— Clarendon Press, Oxford: 57-66.

Kramp, P.L. 1921. Kinetocodium danae, n. g., n. sp. a new gymnoblastic hydroid, parasitic on a ptero-pod.— Vidensk. Meddr dansk naturh. Foren 74:1-21.

Kramp, P.L. 1932. Hydroids. In: The Godthaab Expedition 1928.— Meddr Grönland 79:1-86. Kramp, P.L. 1957. Notes on a living specimen of the hydroid Kinetocodium danae Kramp, parasitic on a

pteropod.— Vidensk. Meddr dansk naturh. Foren. 119: 47-54. Kroiher, M . & G. Plickert, 1992. Analysis of pattern formation during embryonic development of

Hydractinia echinata.— Wilhelm Roux's Arch. Dev. Biol. 201: 95-104. Kurtz, E. & V. Schmid, 1991. Effect of tumor promotors and diacylglycerol on the transdifferentiation

of striated muscle cells on the medusa Podocoryne carnea to RF-amide positive nerve cells.— Hydrobiologia 216-217:11-17.

36 Boero et al. Heterochrony and phylogeny of Hydractiniidae. Zool. Verh. Leiden 323 (1998)

Millard, N . A . H . 1975. Monograph on the Hydroida of southern Africa.— Ann. S. Afr. Mus. 68:1-513. Millard, N . A . H . & J. Bouillon, 1973. Hydroids from the Seychelles (Coelenterata).— Ann. Mus. R. Afr.

Centr., Série in 8°, Sei. Zool. 206:1-106. Motz-Kossowska, S., 1905. Contribution à la connaissance des hydraires de la Méditerranée occiden­

tale. I.- Hydraires Gymnoblastiques.— Archs Zool. exp. Gén. 4: 39-88. Naumov, D.V. 1960-1969. Gidroidi i gidromedusy morskikh, solonovatovodnykh i presnovodnykh

basseinov SSSR.— Opredeleteli po faune SSSR, Izdavaemye Zoologicheskim Institutom Akademii Nauk SSSR. 70 :1-626 (English translation by Israel Program for Scientific Translations, cat. no. 5108, as "Hydroids and Hydromedusae of the USSR").

Petersen, K.W. 1979. Development of coloniality in Hydrozoa. In: G. Larwood & B. Rosen (eds). Biolo­gy and systematics of colonial organisms.— Symp. Syst. Assoc. 11:105-139.

Petersen, K. 1990. Evolution and taxonomy in capitate hydroids and medusae.— Zool. J. Linn. Soc. 100:101-231.

Rasnitsyn, A.P. 1996. Conceptual issues in phylogeny, taxonomy, and nomenclature.— Contr. Zool. 66: 3-41.

Reber-Müller, S., S.-I. Ono, M . Wehrle-Haller & V. Schmid, 1994. Transdifferentiation of striated mus­cle of jellyfish to smooth muscle and nerve cells: the role of cell-ECM interactions and carbohy­drates revealed by a monoclonal antibody.— Differentiation 57: 77-87.

Reber-Müller, S., S.-I. Ono, P. Schuchert, J. Spring & V. Schmid, 1996. Fibrillin in the extracellular matrix of cnidarians: A n immunohistochemical approach.— Sei. Mar. 60: 55-68.

Schmid, V. 1988. The potential for transdifferentiation and regeneration of isolated striated muscle of medusae in vitro.— Cell Differ. 22:173-182.

Schmid, V., H . Alder, G. Plickert & C. Weber, 1988. Transdifferentiation from striated muscle of medusae in vitro. In: G. Eguchi, T.S. Okada & L. Saxen (eds.). Regulatory mechanisms in devel­opmental processes.— Elsevier Scientific Publishers, Dordrecht, Netherlands: 137-146.

Schuchert, P. 1996. Athecate hydroids and their medusae (Cnidaria: Hydrozoa).— N.Z. Oceanogr. Inst. Mem. 106:1-159.

Weber, C. & V. Schmid, 1985. The fibrous system in the extracellular matrix of Hydromedusae.— Tis­sue Cell. 17 (6): 811-822.