Phylogenetic relationships among benthic graptolites
Since the classic work of Kozłowski (1949), the phylogenetic relationships among primitive (i.e. non-graptoloid)
benthic orders of the Graptolithina have only rarely been discussed (e.g. Skevington 1963; Obut 1964; Koz
1962, 1966; Bulman 1968;
Mierzejewski 1985; Urbanek 1986; Chapman et al. 1996). Bulman (1968, 1970) accepted
the validity of only five diverse orders of sessile graptolites (
Camaroidea, Crustoidea, Dendroidea, Stolonoidea,
Tuboidea) and considered that their interrelationships to be unclear. Other graptolite "orders", especially the
Dithecoidea and the Inocaulida, were discussed by Mierzejewski (1986) who considered the majority of dithecoids
(including the family
Chaunograptidae) and inocaulids to be non-graptolitic (presumably colonial scyphopolyps or
hydropolyps, and algae). This view received support, at least in part, from subsequent redescriptions of some
allegedly sessile graptolite genera (e.g. Lo Duca 1990; Mierzejewski 1991).
   Despite the fact that early graptolites are still poorly known, it is possible distinguish two main evolutionary
lineages leading from a common pterobranch ancestor (Kozłowski 1949, 1962): (1)
Crustoidea - Dendroidea -
Graptoloidea; and (2) Tuboidea - Camaroidea. Koz
łowski (1938, 1949, 1962, 1966) stressed the importance of
budding patterns and stolonal morphology in diagnosing the sessile graptolite orders. Hence lineage (1) is
characterized by regular triad budding, whereas lineage (2) exhibits diad budding with no regular succession and
variably distributed nodes. However, Kozlowski left open the question of what mode of budding and thecal
succession has been adopted by the common ancestor of all graptolites. There are some indications that the
Tuboidea and the Camaroidea retained the primary pattern inherited from their pterobranch ancestors,
presumably related to the
Rhabdopleuroidea (Urbanek 1986). Mierzejewski (1985) supposed that the common
ancestor of both lineages must have been very similar, or even closely related, to the mysterious genus
Maenniligraptus Mierzejewski, 1985, which exhibits the both diad and triad modes of budding, and represents one
of the lowest grades of colonial development. However, the problem is made more difficult and complex by some
incertae ordinis graptolites, e.g. genera Mastigograptus Ruedemann, 1908, and Micrograptus Eisenack,
1974 (see Urbanek 1986). Moreover, Skevington (1963) has even suggested that the
Tuboidea was ancestral to the
Denroidea, or
vice versa.
   Until recently, little attention has been paid to the Tuboidea - Camaroidea lineage. These graptolites are poorly
known in terms of their general morphology and astogeny, and their palaeogeographic and stratigraphic distribution.
Their outline morphology was established by Kozłowski (1938, 1949): both are encrusting forms (although some
tuboids have more or less dendroid rhabdosomes), exhibit irregularly diad budding, and possess autothecae which are
strongly differentiated into two portions - a creeping part embedded in a 'thecorhiza' (in tuboids) or 'extracamaral
tissue' (in camaroids), and an erect part, often provided with one or two apertural processes. The Camaroidea are
distinguished from the Tuboidea mainly by (i) the stronger differentiation of camaroid autothecae into two (camara
and collum); (ii) their lack of bithecae (with the exception of
Bithecocamara Kozlowski, 1949); (iii) the lack of
autothecal dimorphism; and (iv) the common occurence of extracamaral tissue instead of stolothecae. In other
words, morphological changes in this lineage were dominated by a significant reduction in polymorphism, a looser
integration of the colony, and a broadening of the creeping part of each autotheca from the tubular shape of tuboids
to the bulbous camara of most camaroids.
   Kozlowski (1949) ascribed a specific role to the tuboid genus Idiotubus Kozlowski, 1949, in the Tuboidea -
Camaroidea lineage [
Idiotubus is treated here as a junior synonym of Epigraptus Eisenack, 1941 (Mierzejewski 1978;
see also Urbanek 1986). Kozlowski considered
Idiotubus to unite certain characters of both orders and to represent a
truly transient link, comparable with the role played by
Dictyonema Hall, 1851 in the Dendroidea - Graptoloidea
lineage (see Kozlowski, 1949, p. 109).
Epigraptus is especially similar to the camaroid genus Graptocamara Kozlowski,
1949. Both have autothecae which arise singly from the surface of a thecorhiza. Moreover, the autothecal apertures
Graptocamara possess a distinct, linguiform process, very similar to that of numerous epigraptid species. The main
difference between the two genera concerns bithecae, which are abundant in the tuboid but absent in the camaroid.
   The encrusting graptolite  Camarotubus graptocamaraeformis Mierzejewski, 2001 seems to represent a stage
morphologically intermediate between the
Tuboidea and the Camaroidea. It departs in some respects from typical
representatives of both orders, while simultaneously combining in mosaic fashion some of their key characters. In the
few well-studied tuboid genera, numerous bithecae outnumber autothecae by several times. As a rule, each tuboid
autotheca is surrounded by a few (2-3 or sometimes even 5) bithecae, forming a more or less regular circle. In
contrast, all known camaroid graptolites (with the exception of the enigmatic
Bithecocamara Kozlowski,1949) are
completely devoid of bithecae. In this respect,
Camarotubus is clearly intermediate in position between the two
orders. Its bithecae are few in number distributed irregularly over the surface of the thecorhiza; this is an important
difference from typical tuboids. However, the basal part of an autotheca in
Camarotubus is distinctly tuboid in form,
i.e. creeping and elongated. Yet its erect portion is under-developed, as in the majority of Camaroidea (with the
exception of
Tubicamara Kozłowski, 1949), and strikingly resembles the collum of the poorly known camaroid species
Graptocamara hyperlinguata Kozłowski, 1949. Remarkably, the autothecal apertures of Graptocamara are sometimes
occluded, as in numerous representatives of the Tuboidea and the Camaroidea (cf. Koz
łowski 1949).
   It can be stated with certainty that  Camarotubus intergrades from the Tuboidea to the Camaroidea. However, while
morphologically intermediate, it cannot be considered a truly transient link because of its age, as is also the case for the
Epigraptus Eisenack and the camaroids Bithecocamara Kozłowski, Graptocamara Kozłowski, and Tubicamara
łowski. It is notable that intermediate forms between the two orders are relatively numerous in comparison with the
small total number of tuboid and camaroid genera. These intergradations make it impossible to define a sharp boundary
between both orders. Strikingly, individual genera exhibit different tuboid and camaroid features. For example,
Epigraptus (=Idiotubus), the typical tuboid graptolite, might playfully to be described as comprising the camara and
apertural apparatus of
Graptocamara, the collum of Tubicamara, and the bithecae of Bithecocamara (which is also
devoid of extracamaral tissue); it certainly lacks the sclerotized stolons of camaroids, but then some other tuboids do
possess this feature. The occurence of different tuboid and camaroid characters intermediate forms suggests that there
were several different evolutionary pathways within the Tuboidea - Camaroidea lineage, as has been shown for the
Dendroidea - Graptoloidea lineage. Moreover, it seems clear that there was a certain independence of particular
rhabdosome elements in the evolution process of the Tuboidea.
  The presence of well-sclerotized stolons in Camarotubus is of special interest because of the few tuboid genera
Tubidendrum Kozlowski, 1949, Kozlowskitubus Mierzejewski, 1978, and partly Reticulograptus Wiman, 1901) which
possess a more or less developed stolonal system with peridermal envelopes (Kozlowski 1949, 1963, 1971; Bulman &
Rickards 1966; Bulman 1970;
Mierzejewski 1978). The significance of these observations for tuboid evolution remains
unclear. Urbanek (1973) pointed out that Kozlowski's (1949) graded series, illustrating successive changes in tuboid
spatial organization, are not necessarily closely relate because of differences in the degree of sclerotization of the
stolon. Later, he suggested that sclerotization of the stolon in the Tuboidea was a gradual process (Urbanek 1986;
see also Bengtson and Urbanek 1986, p. 308). In such a situation,
Epigraptus should be interpreted as distinctly less
advanced than
Camarotubus, in spite of both representing the same morphoecological type and both exhibiting similar
surface features.
Epigraptus may also be interpreted as a morphological intermediate between tuboid graptolites and
some rhabdopleurid pterobranchs (e.g. Middle Cambrian
Rhabdotubus Bengtson et Urbanek, 1986, or 'Rhabdopleura'
Durman et Sennikov, 1993), while Camarotubus bridges the Tuboidea and the Camaroidea (and is especially allied
with early Ordovician
Graptocamara Kozlowski).
The discovery of Camarotubus supports Kozlowski's (1949) view of a close phylogenetic relationship between tuboid
and camaroid graptolites. But paradoxically, the existence of the above mentioned morphological intermediates does
not necessarily confirm Kozlowski's (1949) simple evolutionary pathway from the Tuboidea to the Camaroidea. The
present writer considers the recant data sufficient to cast doubt upon Kozlowski's scheme, and perhaps reject it
altogether. Some Ordovician hemichordates combine, in mosaic fashion, characters of cephalodiscid pterobranchs and
camaroid graptolites (Mierzejewski
1984, 2000 and unpublished results). This may have significant phylogenetic
implications and shed new light on the early evolution of the Graptolithina. The camaroids may, for example, appear not
to be the highly specialized and evolutionarily advanced descendants of the Tuboidea, but even more primitive forms,
closely related to a common ancestor with cephalodiscid-like pterobranchs. These considerations suggest that
Camarotubus may even has been a primitive tuboid, similar to its immediate camaroid ancestors. Nevertheless, the
problem of the origin of the Camaroidea and their phylogenetic relationships is beyond the scope of the present paper
and will be discussed elsewhere (Mierzejewski,
in preparation).
Bengtson, S. and Urbanek, A. 1986. Rhabdotubus, a Middle Cambrian rhabdopleurid hemichordate. Lethaia 19, 293-308.
Bulman, O.M.B. 1968. Graptolithina. Journal of Paleontology  42, 1353.
Bulman, O.M.B. 1970. Graptolithina with sections on Enteropneusta and Pterobranchia. In: C. Teichert (ed.), Treatise on Invertebrate
Paleontology, Part V, Lawrence.
Bulman, O.M.B. & Rickards, R.B. 1966. A revision of Wiman's dendroid and tuboid graptolites. The Bulletin of the Geological
Institutions of the University of Uppsala  
43, 1-72.
Chapman A.J., Durman, P.N. and  Rickards, R.B. 1996. A provisional classification of the graptolite Order Dendroidea. Paläontologische
70, 189-202.
Durman, P.N. and Sennikov, N.V. 1993. A new rhabdopleurid hemichordate from the Middle Cambrian of Siberia. Palaeontology  36,
Kozlowski, R. 1938. Informations preliminaires sur les Graptolithes du Tremadoc de la Pologne et sur leur porte theretique. Annales de
Musei Zoologici Polonici
13, 183-196.
Kozlowski, R. 1949. Les Graptolithes et quelques nouveaus groupes d'animaux du Tremadoc de la Pologne.- Palaeontologia Polonica 3,
Kozlowski, R. 1962. Crustoidea - nouveau groupe de Graptolithes. Acta PalaeontologicaPolonica 7, 3-52.
Kozlowski, R. 1963. Le dĂ©velopment d'un graptolite tuboide.  Acta Palaeontologica Polonica  8, 103-134.
Kozlowski, R. 1966. On the structure and relationships of graptolites.  Journal of Paleontology  40, 489-501.
Kozlowski, R. 1971. Early development stages and the mode of life of graptolites. - Acta Palaeontologica Polonica 16, 313-343.
Lo Duca, S.T. 1990. Medusaegraptus mirabilis Ruedemann, as a noncalcified dasyclad alga. Journal of Paleontology 64, 469-474.
Mierzejewski, P. 1978. Tuboid graptolites from erratic boulders of Poland. Acta Palaeontologica Polonica 23, 557-575.
Mierzejewski, P. 1985. New aberrant sessile graptolites from glacial boulders.  Acta Palaeontologica Polonica 30, 191-199.
Mierzejewski, P. 1986. Ultrastructure, taxonomy and affinities of some Ordovician and Silurian organic microfossils. Palaeontologia
47, 129-220.
Mierzejewski, P. 1991. Estoniocaulis Obut et Rotsk, 1958 and Rhadinograptus Obut, 1960 are not graptolites. Acta Palaeontologica
36, 77-81.
Mierzejewski, P. 2000. An aberrant encrusting graptolite from the Ordovician of Estonia.  Acta Palaeontologica Polonica  45, 239-250.
Obut, A. M. 1964. Podtip Stomochordata. Stomokhordovye. In: Yu.O. Orlov (ed.), Osnovy paleontologii: Echinodermata, Hemichordata,
Pogonophora i Chaetognatha, 279-337. Nedra Press, Moskva.
Obut, A.M. and  Sobolevskaya, R.F. 1967. Nekotorye stereostolonaty pozdnego kembriya i ordovika Norilskogo rayona. In: Novye dannye
po biostratigrafii nishnego paleozoya Sibirskoy platiformy, Akademia Nauk SSR, Sibirskoye otdelenye, Institut geologii i geofiziki, 45-64.
Skevington, D. 1963. Graptolites from the Ontikan Limestones (Ordovician) of Ă–land, Sweden. I: Dendroidea, Tuboidea, Camaroidea, and
Bulletin of the Geological Institutions of the University of Uppsala 42, 1-62.
Urbanek, A. 1973. Organization and evolution of graptolite colonies. In: Boardman, R.S., Cheetham, A.H., and Oliver, W.A. Jr. (eds): Animal
colonies, 441-514. Dowden, Hutchison & Ross, Inc., Stroudsburg.
Urbanek, A. 1979. Ultrastructure of the sicula in the tuboid graptolite Kozlowskitubus erraticus (Kozłowski, 1963). - Acta Palaeontologica
24, 493-503.
Urbanek, A. 1986. The enigma of graptolite ancestry. A lesson from a phylogenetic debate. In Hoffman, A. & Nitecki, M.H. (eds.):
Problematic fossil taxa, 184-226. Oxford University Press, New York.
Urbanek, A., Mierzejewska, G., and Mierzejewski, P. 1980. Scanning electron microscopy of sessile graptolites. Acta Palaeontologica
25, 197-212.
Urbanek, A. and Mierzejewski, P. 1982. Ultrastructure of the tuboid graptolite tubotheca. - Paläontologische Zeitschrift  56, 87-93.

Edited by

Piotr Mierzejewski, the Count of Calmont