Comparative Ultrastructure and Organization of the Prismatic Region of Two Bivalves and its Possible Relation to the Chemical Mechanism of Boring

The prismatic region of two bivalve molluscs exemplifies, inits structure and organization, one of the types of differentiatedcalcareous substrates through which boring organisms must penetrate. The oriented inorganic crystals, separated from one anotherby intercrystalline "spaces", are structurally organized intowell defined prisms. The prisms of each bivalve vary in shapeand size and are delineated from one another by electron-lucent,non-calcified regions. The demineralized organic matrix is also structurally organizedinto prisms, delineated from one another by prism sheaths, andan intraprismatic matrix structurally organized into closelypacked sheet-like compartments and subcompartments in whichthe inorganic crystals are deposited. The non-mineralized intercrystalline "spaces" between the individualinorganic crystals of the same or adjacent rows in a mineralizedsection are occupied by the walls of the intraprismalic sheet-likecompartments. Similarly, the non-calcified electronlucent regionsdelineating one mineralized prism from the next are occupiedby the thick prism sheaths. These portions of the organic matrixwhich fail to mineralize completely undoubtedly provide readypathways for the passage of solutes and solvents through thesetissues of highly ordered, densely packed, inorganic crystals.Moreover, the framework of the organic matrix, which fails tomineralize in these heavily calcified, molluscan substrates,may provide the primary, not the secondary source of chemicalattack during boring, for once the sheaths and compartmentssurrounding the crystals are broken down or solubilized, thecrystals are themselves loosened and freed for mechanical removalby shell-penetrating organisms.     

IN VITRO RECALCIFICATION OF ORGANIC MATRIX OF SCALLOP SHELL AND SERPULID TUBES

Organic matrix was isolated from the shell of the bivalve Argopectenirradians by decalcification. The capacity of the matrix toinitiate formation of crystals similar in form and orientationto the crystals of normal shell was investigated. Decalcifiedshell matrix placed in an inorganic recalcification solutioninitiated the formation of elongate crystals in parallel arrangementcorresponding to the parallel orientation observed in the matrixfibers and similar to the orientation of the long crystals innormal shell. The detailed form of the crystals deposited invitro was different from that of the normal shell crystals.Electron diffraction analysis of remineralized matrix demonstratedthat the material was calcite, the mineral of normal shell. In contrast, the calcareous tube of the serpulid Hydroides dianthushas crystals lacking uniform arrangement and a matrix whichdoes not have a well-oriented structure. The decalcified tubematrix was recalcified and the mineral posited showed some evidenceof normal orientation. The results demonstrate that matrices of Argopecten shell andHydroides tube can induce crystal formation in vitro. Sincethe soluble matrix would be expected to be removed during decalcification,the observed in vitro effects apparently involve the insolublematrix. (Received 19 June 1984;     

EVIDENCE FOR CHEMICAL BORING IN PETRICOLA LAPICIDA (GMELIN, 1791) (BIVALVIA: PETRICOLIDAE)

Specimens of the dead coral-boring bivalve Petri-cola lapicidahave been obtained from Thailand and Jamaica. Although formerlyconcluded to be a mechanical borer, examination of the burrowand the shell strongly suggests chemical boring. Two glandslocated in the inner mantle folds around the antero-ventr'alpedal gape are thought to be involved in this, although onemay secrete the calcareous material cemented to the posteriorshell margin Less specialised petricolids are mechanical borers of stiffmuds, shales and calcareous rocks. A few are nestlers, e.g.,Claudiconcha. As has been recently suggested for other familiesof borers, the Petricolidae constitute another example of theevolution of a specialised chemical borer from a less specialisedmechanically-boring ancestor (Received 20 July 1987;     

Ecological Aspects of Some Coral-Boring Gastropods and Bivalves of the Northwestern Red Sea

More than 17 molluscan species were obtained from burrows incoral substrata at Al-Ghardaga (Hurghada on maps) on the RedSea coast, six of which in particular bore into livingcolonies.The species reported in this paper belong to the families Mytilidae,Coralliophilidae, and Gastrochaenidae. The direction of boringin living corals is to the outside, the borers keeping pacewith the growing coral layer to maintain their burrows open.Coral growth is generally of a higher rate than that of borers,and burrows are accordingly mostly much larger than their inhabitants.There is evidence in such cases that burrows form initiallyby growth of coral around the settling young. Boring of Lithophagaspecies is mostly due to the abrasive action of the shell whichmoves straight and posteroventrally without any rotation. Incoralliophilids,boring is also executed mechanically by the turning movementsof the shell. Boring in dead coral is directed inwards, andburrows are nearly as large as the borers. The latter avoidthe blocking of their burrows (e.g., by a living coral incrustation)either by great siphonal extension (Rocellaria) or the freeends of the shell may be strengthened to maintain the capabilityof boring in the opposite direction (Lithophaga laevigata).Both L. luevigata and Modiola chmamomeus bore mainly mechanicallyby the rocking movements of the shell. Chemical boring is stilla possibility,particularly in the posterior narrow region ofburrows of Modiola lodging the extended pallial siphons whichare deprived of any effective mechanical devices for boring.Therole of boring algae in rarefying bored coral material hasalso to be included as an indirect chemical factor.     

Excavation of Boreholes by the Gastropod, Urosalpinx: An Analysis by Light and Scanning Electron Microscopy

Penetration of shell by the muricid gastropod, Urosalpinx cinereafollyensis, is accomplished by successive alternating periodsof (a) chemical activity by the accessory boring organ (ABO), and (b) rasping by the radula. This paper reports on the functionsof the radula and of the ABO in producing the characteristicgeometry of the borehole, andon the effects of radular teethand of the ABO secretion on the microscopic anatomy of the surfaceof the borehole during the process of shell-boring. Radulae of U. c. follyensis and the surfaces of incomplete boreholesin the shell of Crassoslrea virginica, Mytilus edulis, and Myawere examined by means of light and scanning electron microscopy.Hardness tests of radular teeth andshell of prey demonstratedthat marginal teeth are harder than rachidian teeth, and thatthe range of hardness of rachidianteeth overlaps that of thethree species of shell. Rasping is carried out by two, occasionallythree, of the five rachidiancusps. Rasping patterns are shallowand asymmetric. Rachidian teeth are worn to the base with use;marginal teeth wear onlyslightly as they are employed mainlyin feeding. The distance between the tips of rachidian cuspscorresponds with the interval between the parallel cusp tracesrasped by them in shell. During each rasping period, snailsscrape off about 1/10 to 1/5 of the surface of the chemicallytreated area of the bottom of the borehole. Dissolution of shell is accomplished by secretion from the secretorydisk of the ABO. With each application of the ABO,most or allof the radular marks of the previous rasping period are erasedby solution of a thin layer of shell. The pattern of etchingis specific for each of the species of shell studied. In oysterand mussel shell, initial solubilization occurs through theorganic, non-mineralized, prism sheaths, exposing prismaticforms shown by other workers to be distinctive for these species,and then proceeds into the organic-calcareous structure of individualprisms. Etching of Mya shell revealed no fundamental prismaticform. Shell-penetration includes dissolution of both organiccomplexes and CaCO₃ crystals. Shell-boring by this snail is principally a chemical process,and the geometry of the borehole is generally a reflection ofthe morphology of the ABO.     

SIZE, SHAPE AND SHELL MORPHOLOGY IN THE ANTARCTIC LIMPET NACELLA CONCINNA AT SIGNY ISLAND, SOUTH ORKNEY ISLANDS

Comparison of the size, shape and shell morphology in littoraland sub-littoral morphs of the Antarctic limpet Nacella concinnareveal differences in shell morphology which are enhanced bystructural anomalies within the shells of the two types. Infestationof sub-littoral shells by the conchocelis phase of an endolithicalga significantly affects shell density and total chlorophylllevels in the two shell morphs. The surface sculpture of sub-littoralshells is characterised by a series of grooves, the configurationof which closely resembles that of the radular teeth in N. concinna.Limpets utilise the available food supply within the shell matrixof other limpets by grazing the shell material. Epibiotic growthof calcareous algae prevent erosion and preserve underlyingshell layers. In severe cases, where protection is lacking,intraspecific shell grazing may remove parts of the shell exposingthe internal tissues. The Dominican Gull, Larus dominicanus, is a major shore predatorof both shell morphs. Gull middens contain both shell typesbut are dominated by the more accessible littoral shells. Comparisonof living populations and midden assemblages indicates thatsize and shape selection of prey occurs, with pear-shaped limpetsbetween 21 mm and 29 mm in length being taken preferentially. Apparent differences in shell form are induced by physical,biological and behavioural influences. Littoral animals arerobust in nature, resist avian pre-dation and are not extensivelygrazed whereas those of the sub-littoral are not subject tothe same degree of predatory attention but suffer a gradualdepletion of their shallower shell form through a combinationof algal infection and intraspecific shell grazing. (Received 21 February 1990; accepted 5 July 1990)     

Possible Boring Structures of Sipunculids

At the present time there is no experimental evidence whichlinks the supposed boring activities of sipunculids to a specificorgan or structure. Structures which have been speculativelyassociated in the literature with boring are: hooks and spinesof the introvert, cuticular papillae with associated epidermalglands, anterior and posterior horny shields, and anterior calcareousshields. In this review these structures are described as theyoccur in five representative species of sipunculids collectedby the author from calcareous rock in the Indian Ocean or theCaribbean Sea. The five species are: Pliascolosoma antillarumGrube and Oersted, Phascolosoma dentigerum (Selenka and de Man), Paraspidosiphon steenstrupi (Diesing), Lithacrosiphon gurjanovaeMurina, and Cloeosiphon aspergillum (Quatrefages). Localitiesof collections are cited, habitats and burrows are described,and the behavior of the animals as observed in the field andlaboratory is noted. In view of the morphology of the possibleboring structures and in light of observations on habitats andbehavior, the possible roles of the structures in boring activitiesare discussed. Highly organized horny shields are present at the anterior andposterior extremities of thetrunk or Paraspidosiphon steenstrupi,whereas anterior calcareous shields are characteristic of Cloeosiphonaspergillum and Lithacrosiphon gurjanovae. Papillae and epidermalglands are present in all five of the species but these aremost highly developed in Phascolosoma dentigerum and P. antillarum.Of the species considered, only P. antillarum lacks hooks onthe introvert. Because of the position of the animal within the rock with anteriorend directed toward themouth of the burrow, it is assumed thatthe anterior shields and the hooks of the introvert play nosignificant role in the formation of the burrow. However, therigid papillae of the trunk and the thickened posterior shield,if rubbed against the wall of the burrow, presumably could beutilized in the mechanical attrition of the more friable rock,whereas the secretory products of the numerous epidermal glandsmight be implicated in the chemical dissolution of the hardersubstrates.     

THE LIMPET PATELLOIDA NIGROSULCATA ON INTERTIDAL PLATFORMS IN THE PERTH AREA, WESTERN AUSTRALIA

Patelloida nigrosulcata on intertidal platforms in the Perthmetropolitan area live on the backs of shells of living abalone,Haliotis roei. Over 95% of abalone 30 mm have one or more limpets,and there is a close relationship between abalone and limpetdensity. Sexes are usually separate in P. nigrosulcata, butabout 4% of the population in hermaphroditic. The animals spawntwice annually in winter (May–June) and spring (October–November).The reproductive periodicity of P. nigrosulata is compared toother published data on acmaeids. ^*Present address: Australian Institute of Marine Science, PMBNo. 3, Townsville, Queensland 4810, Australia (Received 23 March 1987;     

Phosphatic Shell Formation in Atremate Brachiopods

The atremate brachiopods are unique in that they possess shellsof calcium phosphate. In Lingula adamsi and Gloltidia pyramidata,the shell mineral is (CO₃ + F)-containing apatite and is crystallo-chemicallysimilar but not identical to the mineral francolite. The shellof Glottidia consists of a thin periostracum, a mineralizedthick primary layer, and alternating mineralized layers andless mineralized chitin layers. The basic unit of the crystalsis the spherulite. Proteinaceous and glycosaminoglycan (GAG)matrices are present in the primary and mineralized layers.The GAGS in the chitin layer are morphologically different fromthose of the other layers. The GAGS are intimately associatedwith the apatite crystals. Shell formation appears to be mediated by three different typesof cells in the outer epithelium. The cells primarily involvedin the mineral formation are characterized by many vacuoleswith electron-dense granular inclusions containing Ca, P, andS. The connective tissue at the anterior edge of the mantlealso contains fine granules with Ca, P, and S. Those granulesare considered to be a mineral reserve for shell formation.Some problems of the mechanisms of shell formation are discussed.     

Structure of shells from eggs of kinosternid turtles

Shells from eggs of five species of kinosternid turtle (Sternotherus minor, Kinosternon flavescens, K. baurii, K. Hirtipes, and K. alamosae) were examined with light and scanning electron microscopy. Except for possible differences among species in thickness of eggshells, structure of shells from all eggs was similiar. In general, kinosternid turtles lay eggs having a rigid calcareous layer composed of calcium carbonate in the form of aragonite. The calcareous layer is organized into individual shell units with needlelike crystallites radiating from a common center. Most of the thickness of the eggshell is attributable to the calcareous layer, with the fibrous shell membrane comprising only a small fraction of shell thickness. Pores are found in the calcareous layer, but they are not numereous. The outer surface of the eggshells is sculptured and may have a thick, organic layer in places. The outer surface of the shell membrane of decalcified eggshells is studded with spherical cores which presumably nucleate growth of shell units during shell formation. The shell membrane detaches from eggs incubated to hatching, carrying with it remnants of the calcareous layer. Such changes in shell structure presumably reflect withdrawal of calcium from the eggshell by developing embryos.     

A NEW SPECIES OF LITHOPHAGA (BIVALVIA: LITHOPHAGINAE) BORING CORALS IN THE CARIBBEAN

A new species of Lithophaga is described as a small lithophaginemussel exclusively boring Madracis mir-abilis, M. decactis andM. formosa in Jamaica. The shell, musculature and pallial glandsshow modifications for live coral boring similar to those ofIndo-Pacific species of the genus. However, both the boringand posterior pallial glands are more primitive than other speciesexamined to date, interpreted as indicative of a more recentadaptation to life in a living coral habitat by this species. ^*Contribution No.359 of the Discovery Bay Marine Laboratory,University of the West Indies (Received 23 April 1985;     

The Morphology, Growth and Reproduction of Unionidae (Bivalvia) in a Fenland Waterway

The shell morphology and population dynamics of the five BritishUnionidae are compared within a sympatric population. Pseudanodontacomplanata is distinguished from Anodonta anatina and A. cygneaby the hinge length–shell length relationship; this morphologicaldistinction may serve as a useful tool in the identificationof this threatened species. The shell length at a given annuluswas remarkably similar for all five species, although the asymptoticlength is reached most quickly in P. complanata and Unio pictorum.P. complanata is relatively short-lived and attains the lowest maximumlength, while A. cygnea lives more than twice as long and attainsalmost double the length of P. complanata. Unio spp. have ashort gravid season over the summer, while Anodonta spp. havea long gravid period, lasting from Autumn through to Spring.Unlike other members of the Anodontinae, P. complanata has ashort breeding season, overlapping with that of the Unio spp. (Received 4 March 1998; accepted 23 April 1998)     

PREDATION ON THE BIVALVE CHOROMYTILUS MERIDIONALIS (KR.) BY THE GASTROPOD NATICA (TECTONATICA) TECTA ANTON

The effects of a population of the boring gastropod Natica tectaon the bivalve Choromytilus meridionalis were investigated atBailey's Cottage, False Bay, South Africa. In July 1979 theN. tecta density on the mussel bed averaged 69 m^–2 andthe population consisted mainly of reproductively mature individualsbetween 20–33 mm shell width. Laboratory experiments on N. tecta showed that prey size selectionis an increasing function of predator size. The prey size rangetaken by large N. tecta is also greater than that taken by smallindividuals. The position of the borehole on the mussel shellis a function of the way in which the shell is held by the footduring the boring process. Consumption rates measured in thelaboratory showed an increase from approximately 1 kJ per weekin 18 mm N. tecta to 4.5 kJ per week in 28 mm individuals. Populationconsumption in the field was calculated as 663 kJ m^–2month^–1. It was estimated that at this rate the standingcrop of mussels in the pool would be eliminated within 10 months.Field measurements showed significant depletion after 6 months. New spat settlement of mussels occur every 4–6 years.The growth curve shows that after one year the population meansize exceeds 30 mm shell length, which is beyond the prey selectionsize range of small N. tecta. It was concluded that at the timeof a new mussel settlement a niche is provided for the simultaneoussettlement and growth of juvenile N. tecta in high densities.However, within one year the increase in prey size, togetherwith depletion due to over-exploitation, limits population growthand density in N. tecta. (Received 14 March 1980;     

A New Ichnospecies of Gastrochaenolites Leymerie from the Pleistocene Port Morant Formation of Southeast Jamaica and the Taphonomy of Calcareous Linings in Clavate Borings

The ichnospecies Gastrochaenolites pickerilli isp. nov. is based on ten borings found in a shell of the gastropod Strombus gigas Linné from the Pleistocene (Sangamonian) Port Morant Formation of southeast Jamaica. These borings bear morphological similarities to Gastrochaenolites torpedo Kelly and Bromley but differ from all other Gastrochaenolites ispp. in having prominent and numerous calcareous meniscate structures arrayed adjacent to one side of the boring. These menisci are concave towards the center of the boring and are the remnants of calcareous tubes that lined earlier boreholes, that the boring bivalve treated as part of the lithified substrate when relocating. They are thus evidence of the former positions of borings that, unusually, were breached as the bivalve migrated sideways. Although this was a common behavior for Gastrochaenolites-producing bivalves within this substrate, the reason for it occurring is uncertain.     

<Emphasis Type="Italic">Polydora uncinata</Emphasis> (Polychaeta: Spionidae) in Chile: an accidental transportation across the Pacific

A Polydora species was found boring in shells of the abalone Haliotis discus hannai cultivated in land-based tanks in Coquimbo, Chile. Spionid polychaetes of Polydora and related genera have been reported from Chile but no worms similar to those found in abalone have been described. The abalone pest corresponds in morphology to Polydora uncinata Sato-Okoshi, 1998, a shell-boring species which was originally described from Japan and never reported from outside the country. It is suggested that occurrence of the species in Chile resulted from its accidental transportation from Japan. Adult worms were most likely transported to Coquimbo with imported abalone brood stock. Prevalence of abalone infestation by worms in Coquimbo varied substantially among cultivation tanks, reaching values as high as 98.8%. Up to 42 worms were found in one shell. The worms often caused formation of nacreous blisters which covered up to 50% of the inner shell surface. Egg capsules with developing larvae were present in female burrows. Larval development was entirely lecithotrophic, with larvae feeding on numerous nurse eggs, staying inside egg capsules until 16–17-segment stage and hatching shortly before metamorphosis. Polydora uncinata is redescribed based on individuals from Coquimbo to alert zoologists in case of accidental release of worms into Chilean coastal waters. Regardless of how the species was transported to Chile, its release to the natural ecosystem may have negative unforeseen impacts on the native fauna.     

Shell disease: abnormal conchiolin deposit in the abalone Haliotis tuberculata

Shell disease in the abalone Haliotis tuberculata L. is characterized by a conchiolin deposit on the inner surface of the shell. The gross clinical signs appear similar to the Brown Ring Disease (BRD) of clams. BRD has been extensively described in clams and is known to be responsible for severe mortalities and the collapse of the clam aquaculture industry in western France. In the clam, it was found to be caused by the infection of the mantle by Vibrio tapetis. Brown protein deposits have been observed in various abalone species around the world; some of these have been associated with a fungal infection in New Zealand, but the ones described here are similar to bacterial infections observed in clams. Larger animals appeared to be more affected by the disease, and a positive correlation of the number of successive infections found in the shells with the level of infestation of the shell by borers suggests that boring polychaetes and sponges may be vectors of the disease, or that the parasite infestation may increase the susceptibility of the animal to this infection. There is no evidence, however, that this infection causes mortality in abalone.     

Early Shell Formation During Molluscan Embryogenesis, with New Studies on the Surf Clam, Spisula solidissima

Shell formation in molluscs begins early in embryogenesis duringsome stage of archenteron formation. Ultrastructural informationon early formation of external shells is available from onlya few bivalves and gastropods. Secretion of the very first shellmaterial by shell field epithelial cells is preceded by an invaginationof the dorsal ectoderm in the region of the shell field. A centuryago, this invagination was termed the "shell gland." As a secretoryfunction for this invagination has not yet been demonstratedand as the term "shell gland" has taken on various meaningsin the literature, the invagination will be referred to as theshell field invagination. The opening into the shell field invaginationseems to be circular in gastropods and elongate in bivalves.Accordingly initial organic shell material seems to form a ringin gastropods and a saddle in bivalves. As in adult molluscs,shells of pre-metamorphic molluscs are composed of both organicand inorganiccomponents. Ultrastructural data from bivalvesand gastropods indicate that the initial organic shell materialis secreted just outside the shell field invagination (acrossthe pore). Initial inorganic shell materials have not been localizednor their pathway traced into or through any pre-metamorphicmolluscs. New SEM and TEM data show that the invagination inthe bivalve Spisula solidissima is composed of a wide outerregion and very narrow inner region with the first shell materialforming at the junction between the two. This is unlike ultrastructuraldata available for other species. Many sections give the falseimpressions that: 1) the shell field invagination is closedto the outside and, 2) that the first organic shell materiallines the innermost region of the invagination. It is not clearwhether the cells of the outer invagination in this speciesare shell field cells. It is suggested that they are not.     

An ultrastructural study of the mantle of the barnacle, Elminius modestus Darwin in relation to shell formation

The fine structure of the mantle and shell of the barnacle, Elminius modestus Darwin has been examined by electron microscopy. The epithelial cells along the outer face of the mantle differ in size, shape, and organelle complexity according to the different components of the shell they secrete. The shell consists of a non-calcareous basis and calcareous mural and opercular plates which are connected by a flexible opercular hinge. Both the basis and opercular hinge are composed of two main units: an outer cuticulin layer and a lamellate component of well ordered arched fibrils. During the deposition of the latter structures morphological changes in the cells occur which may be correlated with the moulting cycle. Preliminary results show that the calcareous plates are covered by an outer epicuticle, which is bordered by a cuticulin layer; the inner calcareous component, consists of an orderly arrangement of organic matrix envelopes within which crystals may be initiated.

The cells lining the inner surface of the mantle are uniform in appearance with a thin cuticle at their free surface which lines the body cavity. The latter structure of the cuticle and manner of its deposition are similar to those of the basis and opercular hinge. Separating the outer and inner mantle epithelial cells is connective tissue which comprises several differing cell types. The possibilities are discussed of the rôle these cells may play in shell deposition. The modes by which underlying cells secrete the different shell components and the cuticle lining the inner face of the mantle, are also discussed.     

Initiation,calcification, and form of larval “archaeogastropod” shells

The coiled shell of gastropods begins as a cap-shaped lens of organic and calcified material that covers the posterior dorsal side of the larva. During development the cap enlarges to cover the larval visceral mass. Marginal growth then produces the characteristic coiled shell. One model of the initiation of shell coiling in “archaeogastropods” requires that the shell remains flexible and uncalcified until after torsion, and that muscle contraction during torsion deforms the shell. We describe early shell calcification and tested this requirement of the model for the patellogastropod limpets Tectura scutum and Lottia digitalis, the trochids Calliostoma ligatum and Margarites pupillus and the abalone Haliotis kamtschatkana. We determined the stage of initial calcification by staining larvae with the fluorescent calcium marker calcein and observing them with bright field, crossed polarizing filter, and fluorescence microscopy. In T. scutum the earliest observable shell was calcified and calcium was sometimes detected even before the initial shell was visible. Larvae of the other species deposited a noncalcified matrix that was subsequently calcified, and in C. ligatum and M. pupillus this initial calcification was distinctly spotty. Shells of both patellogastropods and the abalone were demonstrably rigid prior to torsion while the shells of the trochids were not. These results suggest that shell coiling in patellogastropods and abalone is not initiated by contraction of the larval retractor muscle during torsion; in trochids this mechanism is possible. However, analysis of camera lucida drawings of pre- and post-torsional shells of T. scutum and C. ligatum did not detect shell shape changes during torsion. J. Morphol. 235:77–89, 1998. © 1998 Wiley-Liss, Inc.     

Tube construction in the watering pot shell Brechites vaginiferus (Bivalvia; Anomalodesmata; Clavagelloidea)

The bizarre watering pot shells of the clavagellid bivalve Brechites comprise a calcareous tube encrusted frequently with sand grains and other debris, the anterior end of which terminates in a convex perforated plate (the ‘watering pot’). It has not proved easy to understand how such extreme morphologies are produced. Previously published models have proposed that the tube and ‘watering pot’ are formed separately, outside the periostracum, and fuse later. Here we present the results of a detailed study of the structure and repair of the tubes of Brechites vaginiferus which suggest that these models are not correct. Critical observations include the fact that the external surface of the tube and ‘watering pot’ are covered by a thin organic film, on to the inner surface of which the highly organized aragonite crystals are secreted. There is no evidence of a suture between the tube and the ‘watering pot’ or that the periostracum of the juvenile shell passes through the wall of the tube. Live individuals of B. vaginiferus are able to repair substantial holes in the tube or ‘watering pot’ by laying down a new organic film followed by subsequent calcareous layers. Brechites vaginiferus displays Type C mantle fusion, with the result that the whole animal is encased by a continuous ring of mantle and periostracum, thereby making it possible to secrete a continuous ‘ring’ of shell material. On the basis of these observations we suggest that watering pot shells are not extra‐periostracal but are the product of simple modification of ‘normal’ shell‐secreting mechanisms.