Biosynthetic events of hydrid regeneration

Summary The results of a combined morphological and biochemical study of the role of DNA synthesis during distal regeneration inHydra oligactis revealed that a burst of³H-thymidine incorporation into DNA preceded the elaboration of each of the initial three tentacles. In addition, the relative level of each burst of precursor incorporation relfected the number of tentacles formed at that time. Cytological localization of concentrated amounts of labeled material in nuclei of the hypostome and tentacle regions provided corroborative evidence for the biochemical findings.Evidence that the increased DNA specific activity levels described above are associated with tentacle initiation derived from studies in which regenerating hydra were cultured in hydroxyurea and studies in which hydra regenerated proximally rather than distally. Hydra regenerating in 8 mg/ml (0.105 M) hydroxyurea developed morphologically recognizable hypostomes but no tentacles, and incorporated³H-thymidine into DNA at a level distinctly below that exhibited by uncut, untreated animals. Similarly, hydra regenerated a normal, functional basal disc in the absence of any increased DNA specific activity. Therefore, it is suggested that tentacle initiation inH. oligactis requires concomitant DNA synthesis and, as such, represents an epimorphic phenomenon.     

Head activator does not qualitatively alter head morphology in regenerates ofHydra oligactis

Summary The characterization of head activator (HA) as a morphogen capable of increasing the number of tentacles regenerated by hydra was re-examined. Gastric tissue was excised from HA-treated whole animals and allowed to regenerate. At the cellular level the differentiation of head-specific ectodermal epithelial cells was monitored by quantifying monoclonal antibody, CP8, labeling. This labeling has been correlated with a rise in head activation potential and the determination of tissue to form head structures (Javois et al. 1986). At the morphological level tentacle number was monitored. HA-treated regenerates began the head patterning processes and evaginated tentacles sooner than controls but did not produce extra tentacles. The kinetics of CP8 labeling did not reveal major differences between treated and control regenerates after the initiation of head-specific epithelial cell differentiation. HA appeared to act more like a growth factor stimulating the differentiation of head-specific cell types rather than a morphogen which altered head morphology. An additional aspect of the study examined axial-specific effects of HA on the initiation and extent of head-specific epithelial cell differentiation. The cellular response of ectodermal epithelial cells to HA was dependent on their original axial location. More CP8⁺ tissue differentiated in regenerates of apical as opposed to mid-gastric origin.     

Regeneration in youngBunodactis verrucosa Pennant (actinaria,anthozoa)

Summary YoungBunodactis verrucosa Pennant at the 12 tentacle stage are employed to test the applicability of the polar coordinate model to coelenterate regeneration. The animals are cut along every radius into fragments of 3 to 9 segments. Most fragments are patent 3–4 weeks later, but small fragments have a higher mortality rate than large fragments. Some fragments do not regenerate and occasionally tentacles fuse, thereby reducing the number of segments. Small fragments tend to regenerate more tentacies than large fragments, but large fragments may regenerate great numbers of supernumerary tentacles. Twenty-two percent of the fragments restore the missing number of tentacles, while 76% of all fragments produce an even number of tentacles.Fragments restoring the correct numbers of tentacles show a marked tendency to form the correct tentacles (regulative regeneration). Fragments regenerating two less than the number of tentacles already present show a marked tendency to reproduce tentacles of the types already present (miror image formation). Other fragments produce missing segments (forward regeneration), or those already present (reverse regeneration) at lower frequencies.No fragments beginning or ending with the number 1 directive tentacle fail to regenerate entirely, while first cycle segments maximally remote from segment 1 are associated with the absence of regeneration. No fragments beginning or ending with the number 4 directive tentacle fail to undergo forward regeneration, regulate or produce a mirror image when the appropriate number of segments are regenerated. In contrast, segment 4 is associated with a low frequency of reverse regeneration, and second cycle segments cut away from immediate contact with segment 4 show an increase in the frequency of reverse regeneration. Controls through morphogenic substances rather than polar coordinates seem to explain these results. Such substances would control the number and direction of tentacle regeneration.This work was performed while the author was on sabbatical leave from the University of Pittsburgh at the Stazione Zoologica di Napoli. The author gratefully acknowledges the assistance of Mr. Ciro Gargiulo and of Ms. Gisella Princivalli. This work was supported by a travel grant from the United States Italy Cooperative Science Program of the National Science Foundation. The paper is dedicated to Dr. Alberto Monroy whose generosity made it possible     

VARIATION IN HYDRA'S TENTACLE NUMBERS AS A FUNCTION OF TEMPERATURE

Chlorohydra uiridissima whose tentacle number is altered at different temperatures, was studied to see how other developmental variables changed as a function of temperature. The results suggest that temperature is instrumental in establishing the size of bud and tentacle primordia, but the number of primordia present may play a limiting role.

Animals were cultured at 18, 23 and 28°C and shifted between the extreme temperatures. Large animals with 8 tentacles, small animals with 5 tentacles, and intermediate animals with 6 and 7 tentacles served as parents. Buds and parents were monitored daily and scored for numbers of buds and tentacles.

Temperature, not parental size, determined the size of the buds. At the lower temperature buds were produced more slowly and initiated less frequently, but occurred in greater numbers per parent and had more tentacles than at the higher temperatures. The duration of bud development also increased at lower temperature, but at the lowest temperature the duration of bud development was not correlated with tentacle numbers on buds.

Changes in the frequency of bud initiation and the duration of bud development induced by changing temperature did not parallel changes in the number of tentacles produced on buds. Animals shifted from 18°C to 28°C underwent rapid increases in the rate of bud initiation and rapid shortening in the duration of bud development, while animals shifted from 28°C to 18°C underwent equally rapid changes in the opposite directions. The number of tentacles produced on buds, however, changed slowly to that characteristic of buds acclimated to the new temperatures. The frequency of bud initiation and the duration of bud development, therefore, do not determine tentacle number.

The number of tentacles already present seems to limit possibilities for adding new tentacles. Parents with five tentacles were especially likely to undergo upward changes in their tentacle number while parents with eight tentacles were resistant to such changes.     

Tentacle contraction and ultrastructure inDiscophrya collini: The response to cations

Summary Discophrya collini is a suctorian protozoan with contractile tentacles containing a microtubule-lined canal and microfilaments. The effects of a range of cations on tentacle contraction and ultrastructure have been determined. Treatment with 80 mM CaCl₂ and 95 mM MgCl₂ causes contraction to 28% and 57% of the control length respectively. Re-extension takes over 4 hours in the culture medium, but CaCl₂-treated tentacles are re-extended after a 5 minutes treatment with 10^–2 M EDTA or 5 × 10^–3 M EGTA. CuCl₂ causes a significant contraction at 10^–5 M (to 77%); LaCl₃ at 10^–4 M (to 65%); ZnCl₂ at 10^–2 M (to 65%), but BaCl₂, CoCl₂, MnCl₂, NiCl₂, and SrCl₂ cause significant changes only at 10^–1 M.The cytoplasm of CaCl₂-treated cells contains two forms of membraneous structures when viewed in TEM; that of MgCl₂-treated cells reveals granular areas of medium electron density. None of these features are seen in control cells. The microtubules of the tentacle canal appear to be intact upon its retraction into the cell with no change occurring in the numbers or relative positions of the microtubules. The tentacle cortex is wrinkled. It is suggested from this and previous work that tentacle contraction may be mediated by a microfilament-based mechanism, and that calcium may be involved.     

Electrically induced tentacle contraction in the suctorian protozoonTrichophrya collini

Summary Extracellular electrical stimulation ofTrichophrya collini induces tentacle contraction. There is an inverse relationship between stimulus duration and voltage in producing a threshold response, and at a set voltage the response is graded depending upon duration of stimulus. With a threshold stimulus (6.3 V, 1,000 ms) the response is restricted to the anodal tentacles, and with increasing stimulus intensity or duration the response spreads to the cathodal and finally the intermediate tentacles. With a stimulus of 15 V, 1,000 ms the mean tentacle length is reduced to 28% of the control within 1.2 s. Recordings using intracellular microelectrodes give resting membrane potentials between –10mV and –40mV. Intracellular hyperpolarizing currents of 1nA and 2nA induce tentacle contraction to 50% and 25% of the control length respectively, but depolarizing currents do not induce contraction. SEM studies show that in the initial stages of contraction, only the central region of the tentacle shaft becomes shortened, but on full contraction shortening involves the whole of the shaft. TEM studies show that on contraction no depolymerization of tentacle axoneme microtubules occurs, but that the entire axoneme passes down into the body cytoplasm. These observations are discussed in relation to the possible mechanisms of tentacle contraction.Abbreviations Ax axoneme - C cortex - EDB elongate dense body - SEM scanning electron microscopy - TEM transmission electron microscopy     

Ultrastructural studies of the commensal suctorian,Choanophyra infundibulifera hartog

Summary Tentacle structure, movement and feeding of the commensal suctorian Choanophrya infundibulifera have been examined by light, scanning and transmission electron microscopy. The tentacles possess a flattened tip and rounded shaft externally, with a neck and root region internally. There is a microtubule canal consisting of 150 ring microtubules within which are 20–35 curved lamellae each containing about 20 microtubules. Novel structural features include pairs of short oblique arranged microtubules at the tip, and a collar of epiplasm in the neck region. No haptocysts are found in Choanophrya but the tentacle cytoplasm contains two types of inclusions named solenocysts and spherical vesicles. These features are discussed in relation to the processes of tentacle movement and feeding. The rapid longitudinal movements of the tentacles are described and compared to those of other suctorians and possible mechanisms are suggested. Ingestion in Choanophrya is described and several theories involving tentacle microtubules in the feeding process are examined.This investigation was supported by the J.S. Dunkerley Fellowship in Protozoology, awarded by the University of Manchester.     

Formation of pattern in regenerating tissue pieces of Hydra attenuata: II. Degree of proportion regulation is less in the hypostome and tentacle zone than in the tentacles and basal disc

The relative sizes of the various structures in Hydra attenuata were compared over a broad range of animal sizes to determine in detail the ability to regulate proportions during regeneration. The three components of the head, namely hypostome, tentacles, and tentacle zone from which the tentacles emerge, the body column, and the basal disc were all measured separately. Ectodermal cell number was used as the measure of size. The results showed that the basal disc proportioned exactly over a 40-fold size range, and the tentacle tissue proportioned exactly over a 20-fold size range. In contrast, the hypostome and tentacle zone proportioned allometrically. With decreasing size, the hypostome and tentacle zone became an increasing fraction of the animal at the expense of body tissue, and in the very smallest regenerates at the expense of tentacle tissue. In their current form, the reaction-diffusion models proposed for pattern regulation in hydra are not consistent with the data.     

Epidermal structure of the scleractinian coral Mycetophyllia ferox: light-induced vesicles,copious mucocytes,and sporadic tentacles

Three colony fragments of the scleractinian coral Mycetophyllia ferox Wells from Florida were observed in flow-through seawater aquaria under light and dark conditions. The colonies were then anesthetized and fixed for microscopic examination. Small vesicles formed across the epidermis in response to light as gastrodermis containing approximately 1.9 × 10⁶ zooxanthellae cm⁻² migrated into them. The vesicles flattened in the dark and the gastrodermis retreated to a clumped position. The epidermis is dominated by mucus cells with more than 6300 per mm². In contrast, there are very few epidermal cnidae. The polyps lack tentacles entirely, though small tentacles do occur, albeit sporadically, along the colline walls. Colline tentacles are expanded both day and night, and there is considerable intracolonial variability in the number of cnidae within them, ranging from as few as 316 to more than 3200 per mm² tentacle. There may be several small cnidocyst batteries containing both spirocysts and nematocysts (all microbasic p-mastigophores), but the principal battery is at the tentacle tip where cnidae are much more densely packed. There is considerable variation in the ratio of the two cnidae among tentacles in the same colony. Since the tentacles occur inconsistently and do not appear to expand, their functional role is unclear. Comparisons of epidermal characters are made with other members of the genus Mycetophyllia.     

Formation of pattern in regenerating tissue pieces of Hydra attenuata: I. Head-body proportion regulation

The precision with which an almost uniform sheet of hydra cells develops into a complete animal was measured quantitatively. Pieces of tissue of varying dimensions were cut from the body column of an adult hydra and allowed to regenerate. The regenerated animals were assayed for number of heads (hypostomes plus tentacle rings), head attempts (body tentacles), and basal discs. To ascertain whether the head and body were reformed in normal proportions, the average number of epithelial cells in the heads and bodies was measured. Pieces of tissue, from $\text{1}\text{2}$ to $\text{1}\text{20}$ an adult in size, formed heads that were a constant fraction of the regenerate. Thus, over a 10-fold size range, a proportioning mechanism was operating to divide the tissue into head area and body area quite precisely, but appeared to reach limits at the extremes of the range. However, the regenerates were not all normal miniatures with one hypostome and one basal disc. As the width-length ratio of the cut piece was increased beyond the circumference-length ratio of the intact body column, the incidence of extra hypostomes in the “head” and body tentacles and extra basal discs in the “body” rose dramatically. A proportioning mechanism based on the Gierer-Meinhardt model for pattern formation is presented to explain the results.     

Consequences of food-attraction conditioning in Helix: a behavioral and electrophysiological study

Food-attraction conditioning is a learning phenomenon by which adult Helix pomatia acquire the ability to locate food through exposure to that particular food. Food-conditioned snails can be distinguished from naive snails during their approach to food. Naive snails keep their tentacles upright — whereas food-conditioned animals bend the tentacles down-ward, in a horizontal orientation, pointed in the direction of the food.Tentacle musculature is innervated by two peritentacular nerves (PTn), each projecting to approximately one hemi-section of the tentacle wall. Stimulating the peritentacular nerves caused the tentacles to bend downward in a manner reflecting the full complement of tentacle movements performed by conditioned snails.The neural correlate of tentacle movements was investigated in isolated ganglion preparations with the posterior tentacles attached. PT nerve activity was recorded while the olfactory epithelia were stimulated with natural food odors. Preparations obtained from conditioned animals responded with a substantial increase in unit activity (mean increase 280%) to stimulation with odor of the conditioned food but not to other odors. Preparations from naive animals did not respond to food odor stimulation. The electrophysiological results demonstrated that plasticity due to conditioning the snails in vivo survived dissection and could be monitored in vitro.Abbreviations ext PTn external peritentacular nerve - int PTn internal peritentacular nerve     

Tentacle morphogenesis in hydra

Summary Hydra oligactis exposed to 3 g/ml actinomycin D for 24 hours regenerated only the first pair of tentacles (the mid-laterals). If left uncut, actinomycin D treated animals underwent a reduction of the normal number of tentacles to two or less.Inductive activity was retained in the 2-tentacled hypostomes. However, the tentacles present exhibited reduced capacities to capture and manipulate prey.Histological studies showed that the tentacles of actinomycin D treated hydra were morphologically identical to those of the controls. The interstitial cell (I-cell) population of the treated animals, however, became depleted. Replacing the hypostome of an actinomycin D treated hydra with a normal hypostome reversed the cellular effects of actinomycin D treatment.The modifications in tentacle morphogenesis occurring after actinomycin D treatment are consistent with an impairment of hypostomal function in the animal. It is suggested that the morphological site of this malfunction may be in the nervous system.Research supported by American Cancer Society Institutional Grant No. 342-9157, USPHS Institutional Grant No. 342-9241 and by a grant from Research Corporation.Part of this work was completed while L.H. was an undergraduate research student supported by the National Science Foundation.     

Anchorage and retraction of nematocytes in the tentacles of the cubopolyp Carybdea marsupialis are mediated by a species-specific mesogleal support

Each tentacle of the cubopolyp Carybdea marsupialis is armed with only a single nematocyte at its tip. The correct position of the nematocyte is maintained by a crown-shaped cup formed by the mesoglea. In maximally contracted tentacles, the nematocyte and 7–10 surrounding accessory cells are completely retracted into an ectodermal invagination. A belt of muscle cells revealing a distinct cross-striation in specimens labelled with fluorescein-isothiocyanate-phalloidin is located around the basal part of the nematocyte. These muscle cells are linked both to the mesogleal cup and to the nematocyte by specialized desmosomal contact zones. An additional set of long slender muscle strands runs through the complete length of the tentacles. Their myofibrils reveal only a weak striation pattern. Whereas the contraction of the tentacles seems to depend on the slender muscle strands, the retraction of the apical cell complex is thought to be mediated by the cross-striated muscle belt.     

An IQGAP-related gene is activated during tentacle formation in the simple metazoan Hydra

Differentiation of body column epithelial cells into tentacle epithelial cells in Hydra is accompanied by changes in both cell shape and cell-cell contact. The molecular mechanism by which epithelial cells acquire tentacle cell characteristics is unknown. Here we report that expression of a Hydra homologue of the mammalian IQGAP1 protein is strongly upregulated during tentacle formation. Like mammalian IQGAP, Hydra IQGAP1 contains an N-terminal calponin-homology domain, IQ repeats and a conserved C terminus. In adult polyps a high level of Hydra IQGAP1 mRNA is detected at the basis of tentacles. Consistent with a role in tentacle formation, IQGAP1 expression is activated during head regeneration and budding at a time when tentacles are emerging. The observations support the previous hypothesis that IQGAP proteins are involved in cytoskeletal as well as cell-cell contact rearrangements. Received: 25 January 2000 / Accepted: 2 May 2000     

Ultrastructural studies of the commensal suctorian,Choanophrya infundibulifera hartog

Summary The feeding tentacles of Choanophrya contain a central canal lined by microtubules. Only one tentacle develops during metamorphosis of the embryo into the adult, but others develop at intervals throughout adult life. Each tentacle forms adjacent to a solitary, subcortical kinetosome which lies parallel to the body surface, lacks accessory elements and never develops a cilium. Small condensations of electron-dense material and short bundles of microtubules form adjacent to the cartwheel region of the kinetosome. Initially these bundles are orientated randomly but later they become radially arranged and curved into prolamellae around a disc-shaped condensation centre, to form a paddlewheel-like tentacle primordium 0.8–1.1 m in diameter. The condensation centre consists of alternating concentric electron-dense and electron-transparent zones, and lies with its axis perpendicular to both the kinetosome and the cortex. The microtubules in each prolamella increase in number and pairs of short tip microtubules develop between adjacent prolamellae. Subsequently the developing lamellae become enclosed by a cylinder of ring microtubules. Once all the microtubule components of the tentacle primordium are established it increases in length by addition of material to the basal ends of the microtubules to form a short microtubule canal. As the canal elongates the epiplasm above it disappears and the pellicle membranes become uplifted around the protruding tentacle. An epiplasmic collar differentiates around the growing tentacle whilst spheroid vesicles and solenocysts begin to accumulate in the surrounding cytoplasm.This investigation was supported by the J.S. Dunkerley Fellowship in Protozoology, awarded by the University of Manchester.     

Callus induction and plant regeneration in the metallophyte <Emphasis Type="Italic">Silene vulgaris</Emphasis> (Caryophyllaceae)

Zinc tolerant and non-tolerant ecotypes of Silene vulgaris (Moench) Garcke were examined for their suitability to provide an efficient and reproducible callus formation and regeneration system. Successful and rapid regeneration of adventitious shoots from callus was achieved in leaf tissue but not root or apical meristematic tissue using concentrations of plant growth regulators that spanned a concentration range of (0.05–1 mg l^–1) NAA and (0.5–10 mg l^–1) BAP respectively. Large differences were observed between ecotypes regarding both callus formation and shoot regeneration on the different hormone concentrations. Leaf explants incubated on basal media with different concentrations of auxin/cytokinin demonstrated initial callus formation after 3 weeks of incubation. Callus initiation was seen to develop from the wounded margins of the leaf explants and, after 2 weeks the initially dark callus became more swollen and green. A mean of 6–8 shoots per leaf explant was observed and the survival rate of these regenerates was seen to be 90%. All regenerated plants that were transferred to soil after the emergence of roots, were seen to have no disturbed morphological characteristics. This study demonstrates the stability of the zinc tolerance traits in the regenerated explants and the potential use of this calli formation and regeneration system in Silene vulgaris. Further, this study is a necessary pre-requite for the development of a genetic transformation system with which to study the genetic basis of zinc and, other heavy metal tolerances in a species with a naturally selected high-level tolerance.     

Ontogeny of the fused tentacles in three species of ommastrephid squids (Cephalopoda, Ommastrephidae)

Abstract. The tentacles of ommastrephid squids fuse during embryonic development and remain fused as they grow through hatching, but eventually separate to become two fully functional adult tentacles. The external anatomy of individuals at several post‐hatching ontogenetic stages of three species of ommastrephid squids (Ommastrephes bartramii, Sthenoteuthis oualaniensis, and Hyaloteuthis pelagica) was examined using scanning electron microscopy and morphometrics. The fusion of the transverse muscle mass of the tentacles was examined using light microscopy. Five ontogenetic stages of tentacle separation were defined based on landmark features such as the extent of the fusion and the presence of suckers or sucker buds at the distal tip. The total tentacle length and fused tentacle length reached a maximum when the dorsal mantle length (ML) equaled 3–4 mm (H. pelagica) or 4–6 mm (O. bartramii, S. oualaniensis), and then decreased with increasing ML. The average split length (measured from the base of the tentacles to the point of tentacle fusion) increased gradually with increasing ML, and the separate tentacle diameter was roughly half the diameter of the fused portion at all sizes. In all three species, separation of the fused tentacles began earlier in development (2–3‐mm ML) and was more advanced at smaller sizes than previously reported. The sizes presented here are conservative because excess epithelium at the location of the split may disguise the actual site of separation. Post‐separation tentacles were much shorter than the arms, and the carpal region appeared torn in 2 of the 4 specimens of S. oualaniensis examined. Finally, none of the original distal tip suckers were retained on the post‐separation tentacles of S. oualaniensis. These observations are consistent with the hypothesis that the tentacles separate gradually then rupture at the “wrist” (presumptive carpus), and argue against the possibility of prey capture by the fused tentacles.     

Restoration of morphological and functional integrity in the regenerating eye of the giant African land snail Achatina fulica

To determine whether vision returns to its original state following eye removal in Achatina fulica, light and electron microscope examinations, electrophysiological recordings and behavioural tests were carried out on the regenerating snails. Reparative morphogenesis can result in the restoration of the peripheral sense organ even in the absence of complete regrowth of the tentacle, but it can also lead to the formation of aberrant regenerates. We found that anatomically and ultrastructurally the eyes of the ‘most normal’ regenerates were basically the same as the original eyes. Under normal conditions each eye is composed of a principal and an accessory eye, both sharing a common cornea. The only difference between regenerated and native eyes is the smaller size of the former, as a result of a reduced number of retinal cells. Electroretinographic responses revealed that the molecular mechanism of phototransduction is restored, in principle, but that flicker fusion frequency in the regenerated eye is significantly lower than in the normal eye. The directional movement to a visual stimulus (a black stripe of 45° width) had not completely recovered even 6 months after amputation. This suggests that the central projections of the optic nerve had not become fully re‐established at the time of testing.     

Ultrastructure of tentacles in the sipunculid worm <Emphasis Type="Italic">Thysanocardia nigra</Emphasis> Ikeda, 1904 (Sipuncula)

The ultrastructure of the tentacles was studied in the sipunculid worm Thysanocardia nigra. Flexible digitate tentacles are arranged into the dorsal and ventral tentacular crowns at the anterior end of the introvert of Th. nigra. The tentacle bears oral, lateral, and aboral rows of cilia; on the oral side, there is a longitudinal groove. Each tentacle contains two oral tentacular canals and an aboral tentacular canal. The oral side of the tentacle is covered by a simple columnar epithelium, which contains large glandular cells that secrete their products onto the apical surface of the epithelium. The lateral and aboral epithelia are composed of cuboidal and flattened cells. The tentacular canals are lined with a flattened coelomic epithelium that consists of podocytes with their processes and multiciliated cells. The tentacular canals are continuous with the radial coelomic canals of the head and constitute the terminal parts of the tentacular coelom, which shows a highly complex morphology. Five tentacular nerves and circular and longitudinal muscle bands lie in the connective tissue of the tentacle wall. Similarities and differences in the tentacle morphology between Th. nigra and other sipunculan species are discussed.Original Russian Text Copyright © 2005 by Biologiya Morya, Maiorova, Adrianov.     

Distribution, abundance, behavior and metabolism of Periphylla periphylla, a mesopelagic coronate medusa in a Norwegian fjord

The distribution, behavior and metabolism of the mesopelagic jellyfish, Periphylla periphylla (Péron & Lesueur), were investigated in Lurefjorden, Norway. Field studies, conducted in 1998–1999 with plankton nets and a remotely operated vehicle, indicated that 80-90% of the dense (up to 2.5 m^–3) population migrated 200–400 m vertically each day throughout the year. In situ observations with red light revealed that swimming rates and feeding activity varied with age and time of day. Detection of turbulence and contact with surfaces caused this medusa to conceal one or all of its tentacles in the stomach or to shed nematocyst-laden tissue from the tentacles. Stomachs of medusae collected with nets were often full of prey entangled with the sloughed tissue. Stomachs of medusae captured individually with ROV samplers were empty or contained only a few prey in their stomachs (typically, 1–4 copepods Calanus spp. or chaetognaths Eukrohnia hamata Möbius per medusa). Low rates (0.4–5.6 l O₂ mg C^–1 h^–1) of oxygen consumption of P. periphylla suggested that this species was sustained by relatively few (1–34) prey d^–1.