Figure 1. (A) Plasma membrane (m) separating the cell compartment (c) from the interstitial fluid compartment (e). Symbols: electrical potential of cellular (c) and extracellular fluid (e); chemical activities of j; chemical concentrations of j; activity coefficients of j. (B) Transepithelial pathways in epithelia.

Figure 2. Theories of truly isotonic transport where all of the water exits the epithelium through the interface between the lateral intercellular and the interstitial space and through the opening of the basal plasma membrane infoldings as explained in the the text. (A) In the standing‐gradient theory it is assumed that the lateral Na⁺/K⁺ pumps are displayed near the tight junction and the water channels in the lateral plasma membrane along the entire lateral intercellular space. With appropriate choice of model variables, the condition of zero solute concentration gradient at the interface between the lateral intercellular space and serosal bath is obtained (424). (B) The Na⁺ recirculation theory. The boundary condition of Fig. 2A is replaced by the assumption that a regulated back flux of ions through the serosal plasma membrane, driven by the Na⁺/K⁺ pumps, adjusts the composition of the absorbed fluid to the demanded isotonicity (1301). Note that the Na⁺/K⁺ pumps play a dual role for isotonic transport. Firstly, the pump activity maintains the driving force for water uptake from the luminal solution. Secondly, the pump activity energizes the ion recirculation for achieving an isotonic net transportate. Modified from (1009).

Figure 3. Number of aquatic species in relation to environmental salinity. Redrawn from Remane (1520).

Figure 4. Osmotic concentration in the hemolymph of different crustaceans in ambient media of different osmotic strength. Isosmotic line (dotted line), osmoconformers (bold continuous line), weak hyperosmoregulator (dashed line; values for Carcinus maenas, recalculated after (1660), strong hyperosmoregulators (continuous line; values for Eriocheir sinensis, after [1372]), hyper‐hyposmotic regulator (dotted and dashed line; Artemia salina, values after [360]).

Figure 5. Volume regulation in muscle cells of Chinese crabs (Eriocheir sinensis). Animals acclimated to seawater were transferred to freshwater, and the tissue water content (open circles) and the amino acid concentration (open squares) were monitored over time. Hemolymph and cell osmolality amounted to approximately 1,000 mosmol kg⁻¹ in seawater and dropped to approximately 590 mosmol kg⁻¹ in freshwater. A different group of animals was acclimated to freshwater and then transferred to seawater, and the tissue water content (closed circles) and the amino acid concentration (closed squares) were again monitored. Redrawn after (611), with permission of John Wiley and Sons Ltd.

Figure 6. Independent absorption of Na⁺ and Cl⁻ in the low‐conductance epithelium of posterior gills of the strong hyperosmoregulator Eriocheir sinensis. The figure shows a representative time course of the short‐circuit current (I_SC), demonstrating the negative I_SC (reflecting active, Na⁺‐independent Cl⁻ absorption, further characterized in [1374]) and the positive I_SC (reflecting active, Cl⁻ ‐independent Na⁺‐absorption, further characterized in [2070]) across a split gill lamella. The I_SC deflections are due to 10 mV voltage pulses and are directly proportional to the transepithelial conductance (G_te; see scale on the upper left, redrawn after [2070]). In the presence of high external Cl⁻, most apical Na⁺ channels are closed, and even high concentrations of external Amiloride hardly affect current and conductance. At t = 23 min: In the absence of external Cl⁻ (substitution with gluconate, or at low external Cl⁻; cf. [1372]), apical Na⁺ channels open and external amiloride reduces current and conductance in a concentration‐dependent way. At the end of the experiment, the Na⁺‐dependence of current and conductance is shown by additional substitution of Na⁺ (TRIS). The low conductance at maximal external amiloride is a measure for the paracellular conductance.

Figure 7. Mechanism of transbranchial NaCl absorption in strong hyperosmoregulors, from (556). For further details, see the text.

Figure 8. Coupled absorption of NaCl in the high‐conductance epithelium of posterior gills of the weak hyperosmoregulators Carcinus maenas. The figure shows a representative time course of the short‐circuit current (I_SC), demonstrating the effects of either Na⁺ substitution (choline) or Cl⁻ substitution (gluconate) on both sides of a split gill lamella mounted in a modified Ussing chamber. Current deflections are due to 5 mV voltage pulses and are directly proportional to the transepithelial conductance (G_te; see scale for G_te in lower right, redrawn after [1383]). The larger G_te decrease in the absence of Na⁺ indicates that the paracellular conductance is selective for Na⁺.

Figure 9. Mechanism of transbranchial NaCl absorption in weak hyperosmoregulators (from [556]). For further details see the text.

Figure 10. Schematic diagram of the insect excretory system. Active transport is indicated by solid arrows and passive transport is indicated by open arrows. The flows of the gut contents and the fluid secreted by the Malpighian tubules are indicated by dashed arrows. Adapted from Phillips (1438), with modifications. Abbreviations: organic anions: OAs; organic cations: OCs.

Figure 11. Schematic of the Ramsay assay for measurement of Malpighian tubule fluid secretion rate and secreted fluid solute concentration. An isolated Malpighian tubule is placed in a drop of saline under paraffin oil, and the open end of the tubule is secured to a pin embedded in the Sylgard‐lined bottom of the petri dish. Secreted fluid collects at a puncture in the wall of the tubule between the bathing saline and the pin. Secreted fluid ion concentration ([Ion]) is measured using ion‐selective microelectrodes (1288), atomic absorption spectroscopy (923) or energy‐dispersive X‐ray microanalysis (1902,1998). Organic solute concentrations can be determined using liquid scintillation spectrometry of radiolabeled solutes (1536) or by confocal laser‐scanning microscopic measurement of fluorescence intensity of droplets collected in optically flat glass capillaries (1023). Solute secretion rate (flux) is typically in the femtomol min⁻¹ to picomol min⁻¹ range and is calculated as the product of fluid secretion rate (nl min⁻¹) and solute concentration (μmol l⁻¹ to mmol l⁻¹).

Figure 12. Schematic of ion transport processes involved in fluid secretion by (A) 5HT‐stimulated upper tubules of Rhodnius (822,1342) and (B) Achdo‐KII‐stimulated principal cells in the main segment of cricket Malpighian tubules (316). Dotted lines are used to illustrate the intracellular signaling pathway. Solid circles represent primary active transport processes, whereas open circles are used to represent cotransporters and antiporters. The dark blue symbols represent the receptors for 5HT and for Achdo‐KII and their associated ligands.

Figure 13. Mean values for electrochemical gradients, ion activities and pH in 5‐HT‐stimulated Malpighian tubules of Rhodnius prolixus. Values of pH and ion activity (a_K, a_Cl, a_Na) were recorded 30 min after stimulation with 1 μmol l⁻¹ 5‐HT. Electrochemical gradients (Δμ/F, in mV) for Na⁺, K⁺ and Cl⁻ across the apical and basolateral membranes are taken from (821,824).

Figure 14. Schematic diagram showing net electrochemical potentials (Δμ_net/F) for three cation:Cl cotransporters in serotonin‐stimulated tubules of Rhodnius prolixus. (A) Na⁺‐2Cl⁻‐K⁺, (B) Na⁺‐Cl⁻ and (C) K⁺‐Cl⁻. Based on data in (821).

Figure 15. Regulation of ion transporters by peptides, tyramine and intracellular messengers in Malpighian tubules of Drosophila melanogaster and Aedes aegypti. In Drosophila, the CAPA peptides result in generation of cGMP through the nitric oxide pathway in response to elevation of intracellular Ca²⁺ in the principal cells (908). The calcitonin‐like (CT‐like) diuretic peptide DH31 and the corticotropin releasing factor‐like (CRF‐like) diuretic peptide DH44 lead to increases in intracellular cAMP levels (323,852). Both cAMP and cGMP stimulate the electrogenic V‐type H⁺‐ATPase, leading to an increase in lumen‐positive apical membrane potential (1341). In mosquitoes, the CRF‐like mosquito natriuretic peptide (1437) also results in enhanced Na⁺ entry through a cAMP‐dependent conductive pathway in the basolateral membrane of the principal cells (1611). Kinin receptors are present in the basolateral membranes of stellate cells in tubules of both Aedes and Drosophila (1105,1495). Both the kinins (125,1495,1496,1817) and tyramine (146,147,148) result in increases in intracellular Ca²⁺. In Drosophila, Ca²⁺ is thought to stimulate an increase in transcellular chloride permeability through actions on apical Cl⁻ channels. In Aedes, Ca²⁺ is thought to increase the permeability of the paracellular pathway (126).

Figure 16. Speculative model showing interaction of detoxification and excretion pathways in Malpighian tubules and gut. A moderately hydrophobic toxin (X) may be transported unmodified into the lumen by apical P‐glycoprotein (P‐gp) and/or multidrug resistance‐associated protein (MRP). Alternatively, the toxin may be modified by P450 enzymes and either transported into the lumen by P‐gp, or further modified by conjugation with glutathione (GSH) through the actions of GST and then transported into the lumen by MRP. Toxin X may enter across the basolateral membrane through active (closed circle) or passive (open circle) mechanisms. Adapted from Bard (71,1337).

Figure 17. Schematic representation of accepted (black) and putative (gray) transporters and processes in freshwater fish gill MR cell types (modified from [818]). Classification of cell types is based in part on immunohistochemistry and in part of transport properties of individual cells as assessed by ion‐selective electrodes or isolation of individual cell types. All MR cells express high levels of basolateral Na⁺/K⁺‐ATPase. Rainbow trout PNA⁺ cells express apical NHE isoforms while the PNA⁻ cells express apical H⁺‐pumps and a Na⁺‐channel yet to be identified. Both PNA⁺ and PNA⁻ cells take up Cl⁻ via Cl⁻/HCO₃ ⁻ exchange processes (1402), and although the nature of the anion exchanger(s) involved are unknown, members of the SLC26 family are likely candidates. Zebrafish possess at least three distinct MR cell types. NCC cells are characterized by apical an Na⁺:Cl⁻ cotransporter (NCC) and a basolateral Na⁺:HCO₃ ⁻ cotransporter (SLC4a4) (1028). The NaR cell type, in addition to high levels of Na⁺/K⁺‐ATPase, contains apical and basolateral Ca²⁺ transporters (not shown) and is likely involved in Ca²⁺ acquisition. The third zebrafish cell type, HR, shows strong expression of apical H⁺‐pumps and expression of apical NHE3 as well as basolateral SLC4a1 (1028). Furthermore, cytosolic carbonic anhydrase (CA2) and membrane bound extracellular carbonic anhydrase (CA15) in HR cells are involved in ion uptake. Uptake of Cl⁻ by zebrafish occurs via three members of the SLC26 gene family which have yet to be localized to specific cell types. However, due to the dependence of Cl⁻ uptake on H⁺‐pump activity, the SLC26 transporters have been indicated in NR cells in the present representation. The euryhaline tilapia acclimated to freshwater have as many as four unique MR cell types. So far, type 1 cells have only been demonstrated to express basolateral NKA, and their function remains to be elucidated. Tilapia cell type 2 appears identical to zebrafish NCC cells, with apical NCC and basolateral SLC4a4 cotransport, while cell type 3 express apical NHE3 and basolateral NKCC (presumably NKCC1). Basolateral NKCC(1) is also found in tilapia cell type 4, which also express apical CFTR Cl⁻ channels. It should be noted that the representations illustrate what currently is supported by the peer‐reviewed literature, and that multiple additional transporters and enzyme not included in these diagrams must also be present to account for the observed transport properties of the freshwater teleosts fish gill.

Figure 18. Accepted transport processes in the branchial epithelium of marine teleost fish. MRC = MR cell, AC = accessory cell. Transcellular Cl⁻ secretion includes entry across the basolateral membrane by the secretory isoform of the Na⁺‐2Cl⁻‐K⁺ isoform (NKCC), which is driven by the electrochemical Na⁺ gradient established by the Na⁺/K⁺‐ATPase (∼), and exit across the apical membrane via the cystic fibrosis transmembrane regulator (CFTR) Cl⁻ channel. The active secretion of Cl⁻ establishes a transepithelial potential of −17‐40 mV (reference blood side), which drives paracellular Na⁺ secretion through cation‐selective “leaky junctions” between MRCs and ACs. Cation selectivity of the junctions may be mediated by claudin 10e and possibly other claudins. A basolateral inward rectifying K⁺ channel (eKir) allows for recycling across the basolateral membrane. The depicted transport model also described Na⁺ and Cl⁻ secretion by the renal proximal tubules and the elasmobranch rectal gland. See the text for further details.

Figure 19. Accepted (solid black) and putative (gray) transport processes in the intestinal epithelium of marine teleost fish (redrawn from [652,654,662]). Water transport, transcellular (potentially via AQP1) and/or paracellular (dotted lines) is driven by active NaCl absorption providing a hyperosmotic coupling compartment in the lateral interspace (lis). Entry of Na+ across the apical membrane via cotransporters (NKCC2 and NCC) and extrusion across the basolateral membrane via Na+/K+‐ATPase accounts for transepithelial Na+ movement. In addition, recent evidence that NHE2 and NHE3 are expressed in the intestinal epithelium suggested that Na+ uptake across the apical membrane may occur also via these transporters. Entry of Cl⁻ across the apical membrane occurs via both cotransporters and Cl⁻/HCO₃ ⁻ exchange conducted by the SLC26a6 anion exchanger while Cl⁻ exits the cell via basolateral anion channels. Cellular HCO₃ ⁻ for apical anion exchange is provided in part by HCO₃ ⁻ entry across the basolateral membrane via NBC1 and in part by hydration of endogenous metabolic CO₂. Cytosolic carbonic anhydrase (CAc) found mainly in the apical region of the enterocytes facilitates CO₂ hydration. Protons arising from CO₂ hydration are extruded mainly across the basolateral membrane by a Na+‐dependent pathway and possibly by vacuolar H+ pumps. Some H+ extrusion occurs across the apical membrane via H+ pumps and possibly via NHE2 and/or NHE3 and masks some of the apical HCO₃ ⁻ secretion by HCO₃ ⁻ dehydration in the intestinal lumen yielding molecular CO₂. This molecular CO₂ may diffuse back into the enterocytes for rehydration and continued apical anion exchange. Furthermore, molecular CO₂ from this reaction is rehydrated in the enterocytes and resulting HCO₃ ⁻ is sensed by soluble adenylyl cyclase (sAC), which appears to stimulate ion absorption via NKCC2 (+)(1859). Conversion of HCO₃ ⁻ to CO₂ in the intestinal lumen is facilitated by membrane‐bound carbonic anhydrase, CAIV and possibly other isoforms, a process that consumes H+ and thereby contributes to luminal alkalinization and CO₂ ⁻ formation. Titration of luminal HCO₃ ⁻ and formation of CO₂ ⁻ facilitates formation of CaCO₃ precipitates to reduce luminal osmotic pressure and thus aid water absorption. SLC26a6, the electrogenic anion exchanger, exports nHCO₃ ⁻ in exchange for 1Cl⁻, and its activity is therefore stimulated by the hyperpolarizing effect of the H+ pump. The apical electrogenic nHCO₃ ⁻/Cl⁻ exchanger (SLC26a6) and electrogenic H+ pump constitutes a transport metabolon, perhaps accounting for the apparently active secretion of HCO₃ ⁻ and the uphill movement of Cl⁻ across the apical membrane. The indicated values for osmotic pressure and pH in the absorbed fluids are based on measured net movements of H₂O and electrolytes including H+s, but the degree of hypertonicity and acidity in lis likely are much less than indicated due to rapid equilibration with subepithelial fluid compartments. “TJ” = tight junction. Selective permeability of the epithelium is likely related to multiple isoforms of claudin 3, clauding 15 and claudin 25b, although other claudins may also be involved. See text for further details.

Figure 20. Anatomy of anuran skin (Rana esculenta) with two ion and water transporting units. The multilayered, heterocellular epidermis is an absorbing epithelium of principal cells and MR cells. The acinar epithelium of the subepidermal mucous glands is secretory. Courtesy Dr. Åse Jespersen (993).

Figure 21. Models of the functional organization of the heterocellular anuran skin epithelium. Experimental approaches and methods are detailed in the text. (A) The large Na⁺‐transporting syncytial compartment. Cl⁻ channels are absent in the apical membrane, which contains Na⁺ channels (ENaC) and K⁺ channels. Notably, the Na⁺/K⁺ pumps are located exclusively in the plasma membranes lining the lateral intercellular spaces. (B) The γ‐type MR cell specialized for active and passive Cl⁻ uptake as has been studied in frogs in freshwater and on land covered by saline of regulated composition, respectively. Shown at the apical plasma membrane are voltage‐ and concentration‐gated Cl⁻ channel with its regulatory external Cl⁻ binding site, and the hormone‐gated CFTR like Cl⁻ channel. PKA: protein kinase A; c.s.: catalytic subunit; CA: carbonic anhydrase. As discussed in the text, a number of observations on acid‐base transport and active Cl⁻ uptake are not compatible with the hypothetical γ‐type MR cell. They are, however, reconciled with two other types of MR cells (α and β), originally discovered in turtle urinary bladder. (C) Accepted model of an acid‐secreting α‐type MR cell. (D) Accepted model of a base‐secreting β‐type MR cell with associated non‐rheogenic active Cl⁻ uptake. Modified from (993).

Figure 22. Relationship between Cl⁻ influx (J_in,Cl) and net base excretion (J_net,base) in short‐circuited frog skin (R. esculenta) exposed to 2 mM Cl⁻ on the outside and SO₄ ²⁻ on the inside. Active Na⁺ flux is blocked by amiloride. The regression line is drawn according to The slope of ∼0.5 is compatible with the finding that the anion exchange mechanism also mediates Cl⁻ self‐exchange. Adapted from ref. (480).

Figure 23. Studies by Larsen et al. of the dynamic apical Cl⁻ channels of MR cells of anuran skin. (A) Dependence of the permeability, P_Cl, on Cl⁻ concentration of outside bath at constant V_T (ψ_o − ψ_i) = ‐80 mV (inwardly directed electrical force on Cl⁻); as Cl⁻ _o approaches freshwater concentrations, the channels close reversibly. Adapted from (699). (B) Dependence of the Cl⁻ conductance on V_T at constant [Cl⁻]_o = 110 mM. G_Cl is activated in the physiological range of transepithelial potentials—that is, Cl⁻ channels open when the driving force on Cl⁻ is in the inward direction (V_T < 0 mV) and close when the driving force is outwardly directed (V_T > 0 mV). Adapted from (995).

Figure 24. Flux‐ratio analysis of Cl⁻ transport across the isolated skin of B. bufo. The straight line was calculated by Eq. 14 in the text. The graph shows that the experimental flux ratio follows the theoretical line for electrodiffusion if the driving force, z_Cl·(V_T ‐ E_Cl), is in the inward direction at elevated external Cl⁻, where the apical Cl⁻ channels of the γ‐MR cells are activated (right‐hand side). If z_Cl·(V_T − E_Cl) is in the outward direction (left hand side), the apical Cl⁻ channels are closed while the active Cl⁻ fluxes, fueled by the apical H⁺ V‐ATPase, and the Cl⁻:Cl⁻ exchange diffusion fluxes become the dominating modes of Cl⁻ transport. Thus, the Cl⁻ flux is always inwardly directed, whether the animal is in freshwater of low Cl⁻ or on land and covered by a cutaneous surface fluid of high Cl⁻ concentration (993).

Figure 25. Ion pathways of an acinus cell of submucosal glands of frog skin. The rate of secretion is controlled by the activity of luminal Cl⁻ channels, which are regulated by cAMP and intracellular Ca²⁺. The luminal K⁺ channels are Ca²⁺ and depolarization activated and responsible for the relatively high K⁺ of the secreted fluid. The Na⁺ channels may be of importance for regulating the tonicity of the primary secretion. The model compiles results of pre‐steady state flux‐ratio analyses (1892), whole‐cell and single‐channel patch clamp studies (1733,1734,1735,1736) and investigations of signaling pathways with fluorescent imaging techniques (673). Modified from (1733).

Figure 26. Osmolality of the cutaneous surface fluid (CSF) of R. esculenta (1001, 1002). Number of samples indicated by n is shown for each animal analyzed. Above: CSF samples from animals during evaporative water loss (EWL) at 30‐34°C ranged from 101 to 253 mosmol/kg (mean ± s.e.m., 181±8 mosmol/kg). Below: The osmolality of CSF samples from isoproterenol‐treated frogs at 15 and 23°C ranged from 123 to 281 mosmol/kg (mean±sem, 191±9 mosmol/kg). Horizontal bars indicate lymph concentrations of 10 frogs, 249±10 mosmol/kg (range, 228‐330 mosmol/kg).

Figure 27. Proposed functional coupling of ion and water transport by subepidermal glands and epidermal epithelium of anuran skin; after E.H. Larsen (993). Two MR cells are shown in grey. On land, the skin is kept moist by the cutaneous surface fluid (CSF) produced by the mucosal glands. Water evaporates from the surface fluid (EWL), and the surplus of Na⁺ is absorbed by principal cells (active transport) and Cl⁻ by γ‐MR cells (passive transport). The volume of CSF is maintained by the balance between fluid secretion by subepidermal glands and water reabsorption by solute coupled fluid transport and/or by osmotic water uptake across the surface epithelium. Measured osmolalities are given for the interstitial fluid (lymph) and for CSF during heat‐induced evaporative water loss.

Figure 28. Generalized diagram of amphibian nephron with nomenclature indicated. The proximal nephron comprises Malpighian corpuscle, the ciliated neck segment, the proximal tubule and the ciliated intermediate segment. The subdivision of SI and SII segments is based on dimension of brush border and basolateral infoldings. The Distal nephron comprises the early and late distal tubule, the collecting tubule and the first unbranched portion of the collecting duct system. As shown here, in urodeles and caecilians, the peritoneal funnel connects the proximal nephron with the coelom. In anurans, this connection is lost and the peritoneal funnel opens into the peritubular vessel surrounding the nephron. From (1243).

Figure 29. Simultaneous relationship of afferent arteriolar diameter and blood flow for two representative nephrons in snake (Thamnophis sirtalis) kidney during continuous infusion of arginine vasotocin (AVT). Arrows mark the start of AVT infusion at a rate of 17 pg 100 g⁻¹ min⁻¹. Reproduced from Yokota and Dantzler (2056) with permission.

Figure 30. Model for net tubular secretion of urate based on studies with snake (Thamnophis spp.) proximal renal tubules. Open circle with solid arrow indicates either primary or secondary active transport. For countertransport, solid arrow indicates movement against electrochemical gradient, and broken arrow indicates movement down electrochemical gradient. Broken arrows with question marks indicate possible passive movements. A⁻ indicates anion of unspecified nature. Apparent permeabilities of luminal (P_L) and peritubular (P_P) membranes are shown. Apparent K_t and J_max for net secretion are shown at bottom of the figure. Reproduced from Dantzler and Bradshaw (392) with permission.

Figure 31. Illustration of detailed internal organization of avian kidney. The whole kidney is shown at the lower left with two successive enlargements of sections of the kidney showing increasing internal detail. The largest section shown gives detail of nephrons forming a single medullary cone. Near the surface of the kidney are the small loopless reptilian‐type nephrons arranged in a radiating pattern around a central vein to form a single lobule. In the deeper regions of the cortex are the larger mammalian‐type nephrons with highly convoluted proximal tubules, loops of Henle and distal tubules. A gradual transition occurs from loopless reptilian‐type nephrons to longest‐looped, mammalian‐type nephrons. The loops of Henle are in parallel with collecting ducts and vasa recta within a medullary cone, an arrangement that allows countercurrent multiplication (see text). Reproduced from Braun and Dantzler (187) with permission.

Figure 32. Model for net tubular secretion of urate and PAH in avian proximal tubules, based on studies in chickens. Filled circle with solid arrow indicates primary active transport. Open circle with solid arrow indicates secondary active transport. Open circle with broken arrow indicates carrier‐mediated passive transport. Therefore, for countertransport, solid arrow indicates movement against electrochemical gradient; broken arrow indicates movement down electrochemical gradient. A⁻ indicates anion of unspecified nature. Question marks indicate that these processes have not been clearly demonstrated.

Figure 33. Diagram illustrating the proposed mechanism of secretion in the avian salt gland. Increases in cytosolic Ca²⁺ activate secretion by increasing the opening of basolateral Ca²⁺‐activated K⁺ channels and apical Ca²⁺‐activated Cl⁻ channels. Reproduced from Shuttleworth and Hildebrandt (1684) with permission.

Figure 34. Ammonia‐transporting proteins and ammonia‐transporting mechanisms. Transporters and putative mechanisms are explained in the text. (A) Fictive cell carrying basolateral Na⁺/K⁺‐ATPase (1), apical or basolateral K⁺ channels (2), apical or basolateral NKCC‐cotransporter (3), apical or basolateral cation/H⁺ exchanger (4), apical H⁺/K⁺‐ATPase. (B) Ammonia‐trapping mechanism is energized by H⁺‐ATPase (6). The resulting ΔP _NH3 is driving NH₃ membrane diffusion (7) or facilitated NH₃ diffusion via Rh‐proteins, Amts or MEPs (8). (C) Vesicular ammonia transport based on ammonia trapping and vesicular NH₄ ⁺ transport to target membrane.

Figure 35. Ammonia excretion across the epidermal epithelium in the freshwater planarian Schmidtea mediterranea. The proposed mechanism is described in the text. The figure is modified after (1947).

Figure 36. Branchial ammonia excretion model proposed for the green shore crab Carcinus maenas. The proposed mechanism is described in the text. The figure is modified after (1958).

Figure 37. Ammonia excretion model in the terrestrial ghost crab Ocypode quadrata. The proposed mechanism is described in the text. The figure is modified after (1950).

Figure 38. Ammonia excretion model in the terrestrial crab Geograpsus grayi. The proposed mechanism is described in the text. The figure is modified after (1950).

Figure 39. Branchial ammonia excretion model proposed for the rainbow trout Oncorhynchus mykiss. The proposed mechanism is described in the text. The figure is modified after (1953).

Figure 40. Ammonia excretion model proposed for pavement and mitochondria‐rich cells in the puffer fish Takifugu rubripes. The proposed mechanism is described in the text. The figure is modified after (1953) and (1289).

Figure 41. Branchial ammonia excretion model proposed for mitochondria‐rich cells in the giant mudskipper Periophthalmodon schlosseri. The proposed mechanism is described in the text. The figure is modified after (1953).

Figure 42. Proposed ammonia excretion across the skin of the Mangrove killifish Kryptolebias marmoratus. The proposed mechanism is described in the text. The figure is modified after (1953).