© 2000, Annual Reports of the Zoological Institute RAS.

Evolutionary trends in Crisiidae (Cyclostomata, Bryozoa)

Valentina I. Gontar

Zoological Institute, Russian Academy of Sciences, Universitetskaya nab., 1, St. Petersburg, 199034, Russia

The family Crisiidae belonging to the order bryozoan Cyclostomata have vertical delicate branched calcareous colonies, whose branches are provided with chitinous articulations. Autozooids on the branches may be organized in different ways, depending on which chitinous articulations occur more frequently or less frequently. Autozooids in colonies have a shape of calcareous tubes with walls perforated by numerous small pseudopores. Frequently only the shape of gonozooids, i.e., zooids intended for reproduction helps in the indentification of species. Unlike Cheilostomata displaying essential differences in the autozooidal skeleton between different species and even polymorphism of zooids, Cyclostomata do not possess such features and in the absence of gonozooids it is difficult to identify a species. Skeletal organization is usually accorded major significance in the suborder-level classification of cyclostomids. For resolving such problems Weedon and Taylor (1997, 1998) studied ultrastructure of zooid walls in different families of cyclostomates.

Development of colonies in cyclostomates begins as in other orders of bryozoans. Larva settles and by means of catastrophic metamorphosis the first zooid, or ancestrula, is formed. In Cyclostomata it may be of two types: holoancestrula and artroancenstrula (in crisiids). Further formation of colony by parietal budding occurs. It is important to note that the initial stages of growth in colonies in Cyclostomata maintain a consistent appearance except for crisiids and are called stage of stomatopore, originating from the name of the genus Stomatopora forming such colonies.

In materials from the region of Kuril Islands we have discovered colonies of Vittatopora subrotunda Gontar from Oncousoeciidae with a typical form of colony as in the genus Stomatopora, but with zooids two times smaller. Gonozooid of Vittatopora is pierced by autozooidal peristomes contrasting with the simple globular gonozooid of Stomatopora. It has been established that the ultrastructure of frontal exterior walls of Vittatopora differs essentially from that typical of Tubuliporinae. The most detailed description of different types of ultrastructure are given by Weedon & Taylor, (1997, 1998). Because of limitations in space only their conclusions are used in this paper. The species Anguisia jullieni Ostrovsky was mentioned by Taylor in connection with the study of ultrastructure in Vittatopora. He noted that although resolution of the photographs in the paper by Ostrovsky (1998) is insufficient, Anguisia probably has a similar structure of walls. It can be stated considering literature data that Vittatopora-like pseudopore ultrastructure has been discovered in Vittatopora subrotunda, Tetrastomatopora giselae Moyano, Crisulipora occidentalis Robertson, and probably in Anguisia jullieni. Weedon and Taylor have studied Crisulipora occidentalis (see Fig. 1), and found that it differs from all other articulates (crisiids) in several features and therefore phylogenetic relationships of Crisulipora is a controversial issue.



Fig. 1. Stylized diagrams of frontal exterior walls in a typical articulate (A) and in Crisulipora (B). Walls are progressively stripped to show succession of fabrics comprising the fabric suite. Abbreviations: G - granular fabric; PS - planar spherulitic fabric; TF- transverse fibrous fabric; SN - semi-nacreous, fabric; PF - pseudofoliated fabric after Weedon & Taylor, 1998


They wrote "There is no obvious place for the robust morphology of Crisulipora in Silen's hypothesis, and early fossil history of articulates (Voigt & Walter, 1991) is more consistent with Silen's hypothesis. Silen (1977) has presented a plausible hypothesis for the evolution of articulates from a Stomatopora-like, encrusting uniserial tubuliporine, a model recently supported by Ostrovsky's work on the delicate erect tubuliporine Anguisia (Weedon & Taylor, 1998)".

Further we will discuss in detail the hypothesis of Ostrovsky for the origin of vertical colonies in Crisiidae. The hypothesis of Silen on a probable origin of crisiids from stomatoporiformic ancestors is based on the assumption that peristomal budding in crisiids and stomatoporiformic bryozoans is an identical phenomenon. In reality peristomal budding is widely spread in bryozoans and occurs also in cheilostome and ctenostome bryozoans. If so, the origin of the first crisiids was connected with the transition from encrusting to erect colony growth (Ostrovsky, 1998). The vertical form of colonies occurs among cyclostomates not only in Crisiidae; however this is not accompanied by formation of chitinous joints characteristic of Crisiidae. Further Ostrovsky supposed that "branch strengthening was inseparably linked with the origin of joints in crisiids. Silen's opinion was that jointing "in all probability, is intimately connected with the development of erect growth and thus ought to be regarded as a novelty which appeared recently" (Silen, 1977: 241). In other words, the origin of erect branches preceded the jointing that is generally agreed to be an adaptation for life in localities with strong and irregular water currents." Our studies of ecology of bryozoans permit asserting that under conditions of strong and irregular water currents encrusting bryozoans are predominant, whereas miniature and delicate colonies of Crisiidae occur only in epifaunas of algae and Cheilostomata, i.e. where impact of water flow is almost absent. Further Ostrovsky following Harmelin (1976, 1979) notes that the mode of budding in several fertile branches of Anguisia jullieni is identical to that in Crisiidae and does not occur anywhere else among cyclostomates. Ostrovsky supposed following Borg (1926: 475) "the Crisidia-like colony construction to be ancestral, single-zooid internodes may be a consequence of multiple origin of some kind of structures (in these joints) in ancestral forms (see term 'polymerization', introduced by Dogiel, 1954). Furthermore, data on metamorphosis and early development in crisiids (see Nielsen, 1970) suggests that the acquisition of joints was a secondary process in their evolution. In the first crisiids it may be connected with the occupation of localities with strong irregular water currents. The evolution of more complicated types of structures among crisiids appears to be connected with the increasing of number of autozooids in internodes. The modification of the budding type mentioned above might be an accompanying factor: during the progressive process of compacting of zooidal budding, part of the joints might be lost ('oligomerization' according to Dogiel, 1954). Incidentally, a transition from peristomial budding to budding in the colonial growing zone simultaneously with a gradual transition from one zooidal internode proximally to two-tree zooidal internodes distally can be seen in colonies of Filicrisia (Borg, 1926; Hayward & Ryland, 1985). The gradual complication of colony construction in crisiids was first described by Borg (1926)". In our opinion modification of budding is indirectly correlated with strengthening of branches that were sufficiently firm before origination of articulations in crisiids.

Weedon and Taylor (1998) pointed to polyphyletic evolution of chitinous articulations in bryozoans. The example of Crisulipora as a convergent line of evolution with crisiids is very useful in this respect. Although they wrote that the way how articulations were formed in Crisulipora remains unknown, it can be supposed that they emerge as a consequence of fracture of branches. "Articulations in Crisulipora have an irregular, ragged appearance. The edges of the two internodes on either side of the articulation are not strictly parallel, small patches of calcified wall may form 'islands' between internodes, and the extent of exposed elastic joints varies around the circumference of the joint. Exterior walls are thickened close to joints where the elastic joint was anchored". In recent crisiids "the flattened surface where internodes articulate represent a unique wall surface for cyclostomes. Internal ring-shaped structures are developed which are clearly demarcated being separated by a groove from the inner surface of the wall to which they attached. These ring diaphragms are perpendicular to the wall surface at distal ends of internodes following articulation". As Weedon and Taylor wrote, "Crisulipora differs from all other studied articulates in several respects. Most notably, it has a complex gonozooid pierced by autozooidal peristomes, contrasting with the simple globular gonozooids of other articulates (see above comparison between Vittatopora and Stomatopora - V.G.), and internodes with up to five transverse rows of autozooids, whereas other articulates are biserial or uniserial. Colonies are generally more robust with appreciably larger zooids: autozooid aperture diameter averages about 0.15 mm in C. occidentalis compared, for example, with values of between 0.05 and 0.10 mm in the 10 species of articulates found in the British fauna (Hayward & Ryland, 1985). Gonozooids of Crisulipora further differ from those of other articulates in having non-terminal ooeciopores and lacking the charcteristic 'atrial ring' of Levinsen (1912) (='valve' of Harmer 1893), a structure otherwise known only in meliceritid and some multisparsid fossil cyclostomates (Taylor & Weedon, 1996). The locations of articulation is much less regular in Crisulipora. Clusters of closely spaced internodes often originate from a parent branch, and to quote from Robertson's original description of C. occidentalis (Robertson, 1910:.255), "branching [is] extremely irregular, following no particular method; arising both from lateral and ventral zooecia at any point of an internode…. Branching patterns are sufficiently invariant in other articulates to be useful in distinguishing between different species (e.g. Ryland, 1967, Fig. 5)."

In addition to having jointed colonies, Osburn (1953) pointed out two other similarities between Crisulipora and Crisia: the development of jointed rhizoids or radicles consisting of tubular kenozooids, and the upright growth of the ancestrular tube from the primary disc (protooecium). The former feature is possibly unique to the articulates among cyclostomes, although the latter can also be found in hornerid cancellates.

As a consequence of its atypical morphology, the phylogenetic affinity of Crisulipora has been a matter of debate. Robertson (1910) regarded it as a tubuliporine, and Canu and Basler (1920) even placed it in the tubuliporine family Diaperoeciidae. However, Borg (1926), Osburn (1953), and Soule et al. (1995) have all assigned Crisulipora to the articulate family Crisiidae, although Borg (1926) noted that a thorough investigation of Crisulipora would be highly desirable in view of its tubuliporine-like gonozooid.

Information on skeletal ultrastructure of Crisulipora has been unavailable prior to the current study. The observation of a transverse fibrous layer in both interior and exterior walls of Crisulipora is a major contrast with its absence in all other articulates and favours a closer relationship with tubuliporines in which this fabric is common (Weedon & Taylor, 1997)". Skeletal ultrastructure of Crisulipora is similar to tubiliporines and posses a unique characteristic of pseudopores ultrastructure, found only in Vittatopora subrotunda, Tetrastomatopora giselae, Anguisia jullieni.

Weedon and Taylor (1998) proceeding from the complex of above-mentioned characteristics suggest acknowledging that "elastic joints have evolved twice in cyclostomes, once in Crisulipora from an ancestor with a skeleton incorporating transverse fibres, and again in more typical articulates from an ancestor lacking transverse fibres. Furthermore, the strikingly different ultrastructure of skeletal articulation surfaces is formed of a coarsely porous thickening of skeletal wall, whereas in other articulates it is a smooth ring diaphragm". As noted above the articulations in Crisulipora or ancestors similar to it could have originated as a consequence of branch fracture. This process could have taken place also in crisiids. An argument in favour of this inference is a presence of basis rami in crisiids. Moreover, Crisulipora may be regarded as striking evidence of polymerization processes in the context of Dogiel's theory, which manifest themselves as chaotic multiplication of different structures. Strictly speaking originating of life form is discussed. An analogous plausible evolutionary model for crisiids is proposed. Extending this conclusion it is implied that the hypothesis of Borg-Ostrovsky of an ancestor similar to Crisidia and Filicrisia is alternative. By contrast one-zooidal internodia (as it occurs in Crisiella) is a consequence of oligomerization processes, i.e. ordering and decline of the number of homologous structures (e.g. of the number of zooids in the internodia), which were polymerized at the early stage of evolution of the group. An example of oligomerization process in the genus Crisiella is C. diversa with an only one-zooidal internode and C. chirpoensis described by us with several one-zooidal internodes near the base of the colony.

Weedon (1998) studied ultrastructure of early stages of astogenesis in cyclostomates and came to a conclusion that protooecial ultrastructure is independent of adult structure. "The uniformity of skeletal ultrastructure in cyclostomes corresponds with the close similarity of larvae and post-settlement metamorphosis in the order. The fabric suite of the protoecium resembles the skeletal ultrastructure of Palaeozoic stenolaemates. The primitive fabric condition is retained by some tubuliporines and cancellates. Complex multilayered fabric suites may have evolved in the Mesozoic by addition of new fabric types". It is necessary to add that artroancestrula of Crisiidae (i.e. with articulation) should be regarded as ancestrular complex that originated as a consequence of heterochrony. Analogues have been found in cheilostomates in acquiring ancestrula complexes originating in the process of polyphyletic evolution. Finally, resting on the above-mentioned model, it remains to be noted that different colony shapes in different crisiid genera appear to be correlated with different stages of oligomerization process. It seems that in some cases polyphyletic evolution of crisiid colonies could have taken place. Probably it could have happened that chitinous joints of crisiids having dissaperared once did not evolve again in progressive evolution.


Borg, F. 1926. Studies on recent cyclostomatous Bryozoa. Zool. Bidr. Upps. 10: 181-507.

Canu, F. & R.S. Bassler. 1920. North American early tertiary Bryozoa. Bull. U. S. natn. Mus. 106: 1-876.

Dogiel, V.A. 1954. Oligomerizatsiya gomologichnykh organov [Oligomerization of homologous organs]. Leningrad, LGU Publ. 368 pp. (In Russian.)

Hayward, P.J. & J.S. Ryland. 1985. Cyclostome bryozoans. Synopses Br. Fauna (n. ser.) 34: 1-147. 

Harmelin, J.G. 1976. Les sous-ordre des Tubuliporina (Bryozoares Cyclostomes) en Méditerrannée. Écologie et systématique. Mém. Inst. océanogr., Monaco 10: 1-326.

Harmelin, J.G. 1979. On some stomatoporiform species (Bryozoa, Cyclostomata) from the bathyal zone of the Northeastern Atlantic Ocean. In: Advances in Bryozoology. (G.P. Larwood and M.B. Abbott. Eds.). pp. 403-422. London/New York, Academic Press.

Osburn, R.C. 1953. Bryozoa of the pacific coast of America. Pt. 3: Cyclostomata, Ctenostomata, Entoprocta, and addenda. Allan Hancock Pacif. Exped. 14: 613-841.

Ostrovsky, A.N. 1998. The genus Anguisia as a model of a possible origin of erect growth in some Cyclostomatida (Bryozoa). Zool. J. Linn. Soc. 124: 355-367.

Robertson, A. 1910. The cyclostomatous Bryozoa of the west coast of North America. Univ. Calif. Publs Zool. 6: 225-284.

Ryland, J.S. 1967. Crisiidae (Polyzoa) from western Norway. Sarsia 29: 269-282.

Silen, L. 1977. Structure of adnate colony portions in Crisiidae (Bryozoa, Cyclostomata). Acta zool., Stockh. 58: 227-244.

Soule, D.F., Soule, J.D. & H.W. Chaney. 1995. The Bryozoa. In: Taxonomic atlas of the Santa Maria Basin and Western Santa Barbara Channel. Vol. 13. Los Angeles, University of South California Press. 344 pp.

Taylor, P.D. & M.J. Weedon. 1996. Skeletal ultrastructure and affinities of eleid (meliceritid) Cyclostomate bryozoans. In: Bryozoans in space and time. (D.P. Gordon et al. Eds.). pp. 341-350. Wellington, National Institute of Water and Atmospheric Research Publ.

Voigt, E. & B. Walter. 1991. De possibles Crisiidae (Bryozoa-Articulata) dans le Néocomien du Jura franco-suisse. Geobios 24: 41-46.

Weedon, M.J. & P.D. Taylor. 1997. Skeletal ultrastructure in some Tubuliporine Cyclostome Bryozoans. Acta zool., Stockh. 78 (2): 107-122.

Weedon, M.J. & P.D. Taylor. 1998. Skeletal ultrastructure in some Articulate Cyclostome Bryozoans. Acta zool., Stockh. 79 (2): 133-148.

Weedon, M.J. 1998. Skeletal ultrastructure of the early astogenetic stages of some Cyclostome Bryzoans. Acta zool., Stockh. 79 (2): 163-174.