A new photosensitive pigment of the euryhaline teleost,Gillichthys mirabilis

Retinal extracts have been prepared from dark-adapted mudsuckers by treatment of retinal tissue or of isolated outer segments of the visual cells with digitonin solution. The extracts were examined spectrophotometrically and found to absorb light maximally between the wave lengths of 488 and 510 mµ, depending on the proportion of yellow impurities and light-sensitive pigment present. This photosensitive pigment was shown to be homogeneous by partial bleaching of the extracts with monochromatic light of various wave lengths from 390 to 660 mµ. The mudsucker pigment was specifically demonstrated not to be a mixture of rhodopsin and porphyropsin; the adequacy of the method used to analyze such mixtures was shown by performing a control experiment with an artificial mixture of bullfrog rhodopsin and carp porphyropsin. Comparison of the hydroxylamine difference spectrum and of the absorption maximum of the purest retinal extract located the mudsucker photosensitive pigment maximum at 512 ± 1 mµ. Extraction of retinal tissue with a fat solvent after exposure to white light gave a preparation which after the addition of antimony chloride reagent developed the absorption band maximal near 664 mµ, which is characteristic of retinene₁. If an hour intervened between exposure of the retinal tissue to light and extraction of the carotenoid, the antimony trichloride test gave a color band maximal at 620 mµ, characteristic of vitamin A₁. No evidence of retinene₂ or vitamin A₂ was obtained. The euryhaline mudsucker has, therefore, a photosensitive retinal pigment with an absorption maximum halfway between the peaks of rhodopsins and of porphyropsins and belonging to the retinene₁ system characteristic of rhodopsins. The pigment is therefore named a retinene₁ pigment 512 of the mudsucker, Gillichthys mirabilis. It is uncertain whether this type of photosensitive pigment will be found in other euryhaline fishes.     

THE RATE OF CO2 ASSIMILATION BY PURPLE BACTERIA AT VARIOUS WAVE LENGTHS OF LIGHT

1. The relative absorption spectrum of the pigments in their natural state in the photosynthetic bacterium Spirillum rubrum is given from 400 to 900 mµ. The position of the absorption maxima in the live bacteria due to each of the pigments is: green pigment, 420, 590, 880; red pigment, 490, 510, 550. 2. The relative absorption spectrum of the green pigment in methyl alcohol has been determined from 400 to 900 mµ. Bands at 410, 605, and 770 mµ were found. 3. The wave length sensitivity curve of the photosynthetic mechanism has been determined and shows maxima at 590 and about 900 mµ. 4. It is concluded that the green bacteriochlorophyll alone and not the red pigment can act as a light absorber for photochemical CO₂ reduction.     

The Photosensitive Retinal Pigment System of Gekko gekko

Retinal extracts of Gekko gekko were found to contain two retinene₁ photopigments, one with maximum absorption at about 521 mµ, the second with a maximum in the region of 478 mµ. These pigments were assayed by the method of partial bleaching and their spectral characteristics studied by examining their difference spectra. The 478 mµ pigment was present in the extracts as 8 per cent of the total photopigment concentration. The two pigment systems were shown to be biochemically independent and to have different properties. Unlike the 521 mµ pigment, for example, the 478 mµ pigment was found to resist the action of NH₂OH and, within the cells, to be unaffected by sucrose solutions. These solutions destroyed or altered the 521 system so that extracts of sucrose-treated retinae were found to contain significantly less 521 photopigment. In digitonin solution the 521 pigment was unaffected by sucrose treatment. Both pigments were extracted from separated, washed outer segments and so are considered to be visual pigments. This dual system accounts for the spectral sensitivity of this gecko as determined by Denton. A search was made, but no evidence was secured for the presence of a photopigment absorbing at longer wavelengths. Electoretinographic data suggest, however, that an elevated sensitivity at longer wavelengths occurs in some geckos so that a continued search is justified for a third photopigment.     

The photosensitive retinal pigments of fishes from relatively turbid coastal waters

Digitonin extracts have been prepared from the retinae of a dozen species of marine and euryhaline teleost fishes from turbid water habitats. Spectrophotometric analysis of the extracts shows that the photosensitive retinal pigments of these species have maximum absorption above 500 mµ. In nine species there are retinene₁ pigments with λ_max between 504 and 512 mµ. In the marine but euryhaline mullet, Mugil cephalus, there is a porphyropsin with λ_max 520 mµ. A mixture of rhodopsin and porphyropsin in an extract of a marine puffer, Sphoeroides annulatus, was disclosed by partial bleaching with colored light. In addition, one other species has a 508 mµ pigment, of which the nature of the chromophore was not determined. The habitats in which these fishes live are relatively turbid, with the water greenish or yellowish in color. The spectral transmission of such waters is probably maximal between 520 and 570 mµ. It is suggested that the fishes have become adapted to these conditions by small but significant shifts in spectral absorption of their retinal pigments. These pigments are decidedly more effective than rhodopsin in absorption of wavelengths above 500 mµ. This offers a possible interpretation of the confusing array of retinal pigments described from marine and euryhaline fishes.     

Single and Multiple Visual Systems in Arthropods

Extraction of two visual pigments from crayfish eyes prompted an electrophysiological examination of the role of visual pigments in the compound eyes of six arthropods. The intact animals were used; in crayfishes isolated eyestalks also. Thresholds were measured in terms of the absolute or relative numbers of photons per flash at various wavelengths needed to evoke a constant amplitude of electroretinogram, usually 50 µv. Two species of crayfish, as well as the green crab, possess blue- and red-sensitive receptors apparently arranged for color discrimination. In the northern crayfish, Orconectes virilis, the spectral sensitivity of the dark-adapted eye is maximal at about 550 mµ, and on adaptation to bright red or blue lights breaks into two functions with λ_max respectively at about 435 and 565 mµ, apparently emanating from different receptors. The swamp crayfish, Procambarus clarkii, displays a maximum sensitivity when dark-adapted at about 570 mµ, that breaks on color adaptation into blue- and red-sensitive functions with λ_max about 450 and 575 mµ, again involving different receptors. Similarly the green crab, Carcinides maenas, presents a dark-adapted sensitivity maximal at about 510 mµ that divides on color adaptation into sensitivity curves maximal near 425 and 565 mµ. Each of these organisms thus possesses an apparatus adequate for at least two-color vision, resembling that of human green-blinds (deuteranopes). The visual pigments of the red-sensitive systems have been extracted from the crayfish eyes. The horse-shoe crab, Limulus, and the lobster each possesses a single visual system, with λ_max respectively at 520 and 525 mµ. Each of these is invariant with color adaptation. In each case the visual pigment had already been identified in extracts. The spider crab, Libinia emarginata, presents another variation. It possesses two visual systems apparently differentiated, not for color discrimination but for use in dim and bright light, like vertebrate rods and cones. The spectral sensitivity of the dark-adapted eye is maximal at about 490 mµ and on light adaptation, whether to blue, red, or white light, is displaced toward shorter wavelengths in what is essentially a reverse Purkinje shift. In all these animals dark adaptation appears to involve two phases: a rapid, hyperbolic fall of log threshold associated probably with visual pigment regeneration, followed by a slow, almost linear fall of log threshold that may be associated with pigment migration.     

THE PORPHYROPSIN VISUAL SYSTEM

1. In the rods of fresh-water and some anadromous fishes, rhodopsin is replaced by the purple photolabile pigment porphyropsin. This participates in a retinal cycle identical in form with that of rhodopsin, but in which new carotenoids replace retinene and vitamin A. 2. Porphyropsin possesses a broad absorption maximum at 522 ± 2 mµ, and perhaps a minimum at about 430 mµ. The vitamin A-analogue, vitamin A₂, possesses a maximum in chloroform at 355 mµ and yields with antimony trichloride a deep blue color due to a band at 696 mµ. The retinene-analogue, retinene₂, absorbs maximally in chloroform at 405 mµ and possesses an antimony chloride maximum at 706 mµ. 3. Its non-diffusibility through a semi-permeable membrane, salting-out properties, and sensitivity to chemical denaturants and to heat, characterize porphyropsin as a conjugated carotenoid-protein. 4. The porphyropsin cycle may be formulated: porphyropsin See PDF for Structure. retinene₂-protein ₍₂₎ ^→ vitamin A₂-protein ₍₃₎ ^→ porphyropsin. Isolation of the retina cuts this cycle at (3); denaturation procedures or extraction of porphyropsin into aqueous solution eliminate in addition (1) and (2). 5. The primary difference between the rhodopsin and porphyropsin systems appears to be the possession by the latter of an added ethylenic group in the polyene chain.     

THE PHOTOTROPIC SENSITIVITY OF PHYCOMYCES AS RELATED TO WAVE-LENGTH

Under the circumstances of experimentation described, the sporangiophores of Phycomyces are found to be most sensitive to stimulation by light in the violet between 400 and 430 mµ. Toward the red, sensitivity falls to nearly zero near 580 mµ, while in the near ultra-violet around 370 mµ, sensitivity is still high. The previous experiments of Blaauw had placed the point of greatest sensitivity some 80 mµ nearer the red end of the spectrum. Because of the known presence in the sporangiophores of Phycomyces of "accessory" pigments, care must be taken in identifying such results with the absorption spectrum of the photosensitive substance.     

THE PIGMENT-PROTEIN COMPOUND IN PHOTOSYNTHETIC BACTERIA : II. THE ABSORPTION CURVES OF PHOTOSYNTHIN FROM SEVERAL SPECIES OF BACTERIA

Absorption curves have been obtained in the spectral region of 450 to 900 mµ for the water soluble cell juice of four species of photosynthetic bacteria, Spirillum rubrum (strain S1), Rhodovibrio sp. (strain Gaffron), Phaeomonas sp. (strain Delft), and Streptococcus varians (strains C11 and orig.). These curves all show maxima at 790 and 590 mµ due to bacteriochlorophyll, whose highest band, however, occurs at 875, 855, or 840 mµ depending on the species. The bacteria that appear red rather than brown have a band at 550 mµ due to a carotinoid pigment. An absolute absorption curve of bacteriophaeophytin has maxima at 530 and 750 mµ. The extraction of cell juice by supersonic vibration does not change the position of the absorption bands or of the light absorbing capacity of the pigment.     

The spectral sensitivities of the dorsal ocelli of cockroaches and honeybees; an electrophysiological study

The spectral sensitivities of the dorsal ocelli of cockroaches (Periplaneta americana, Blaberus craniifer) and worker honeybees (Apis mellifera) have been measured by electrophysiological methods. The relative numbers of quanta necessary to produce a constant size electrical response in the ocellus were measured at various wave lengths between 302 and 623 mµ. The wave form of the electrical response (ERG) of the dark-adapted roach ocellus depends on the intensity but not the wave length of the stimulating light. The roach ocellus appears to possess a single photoreceptor type, maximally sensitive about 500 mµ. The ERG's of bee ocelli are qualitatively different in the ultraviolet and visible regions of the spectrum. The bee ocellus has two types of photoreceptor, maximally sensitive at 490 mµ and at about 335 to 340 mµ. The spectral absorption of the ocellar cornea of Blaberus craniifer was measured. There is no significant absorption between 350 and 700 mµ.     

Photoreversal of nuclear and cytoplasmic effects of short ultraviolet radiation on Paramecium caudatum

1. Irradiation with three short ultraviolet (UV) wave lengths, 226, 233, and 239 mµ rapidly immobilizes Paramecium caudatum, the dosage required being smaller the shorter the wave length. 85 per cent of paramecia immobilized with wave length 226 mµ recover completely. Recovery from immobilizing doses is less the longer the wave length. 2. Irradiation continued after immobilization kills the paramecia in a manner which is markedly different for very short (226, 233, and 239 mµ) and longer (267 mµ) wave lengths. 3. An action spectrum for immobilization in P. caudatum was determined for the wave lengths 226, 233, 239, 248, and 267 mµ, and found to resemble the absorption of protein and lipide in the wave length region below 248 mµ. Addition of these data to those of Giese (1945 b) gives an action spectrum resembling the absorption by albumin-like protein. 4. Division of P. caudatum is delayed by doses of wave lengths 226, 233, and 239 mµ which cause immobilization, the longest wave length being most effective. 5. Immobilization at any of the wave lengths tested (226, 233, 239, 248, 267 mµ) is not photoreversible when UV-treated paramecia are concurrently illuminated. 6. Division delay resulting from immobilizing doses of 226, 233, and 239 mµ is photoreversible by exposure to visible light concurrently with the UV. 7. Division delay induced by exposure to wave length 267 mµ is reduced by exposure to visible light applied concurrently with UV or immediately afterwards. 8. The data suggest that the shortest UV wave length tested (226 mµ) affects the cytoplasm selectively, because it is absorbed superficially as indicated by unilateral fluorescence in UV. Consequently it immobilizes paramecia rapidly but has little effect on the division rate because little radiation reaches the nucleus. 9. The data support the view that nuclear effects of UV are readily photoreversed but cytoplasmic effects are not.     

The interplay o light and heat in bleaching rhodopsin

Rhodopsin, the pigment of the retinal rods, can be bleached either by light or by high temperature. Earlier work had shown that when white light is used the bleaching rate does not depend on temperature, and so must be independent of the internal energy of the molecule. On the other hand thermal bleaching in the dark has a high temperature dependence from which one can calculate that the reaction has an apparent activation energy of 44 kg. cal. per mole. It has now been shown that the bleaching rate of rhodopsin becomes temperature-dependent in red light, indicating that light and heat cooperate in activating the molecule. Apparently thermal energy is needed for bleaching at long wave lengths where the quanta are not sufficiently energy-rich to bring about bleaching by themselves. The temperature dependence appears at 590 mµ. This is the longest wave length at which bleaching by light proceeds without thermal activation, and corresponds to a quantum energy of 48.5 kg. cal. per mole. This value of the minimum energy to bleach rhodopsin by light alone is in agreement with the activation energy of thermal bleaching in the dark. At wave lengths between 590 and 750 mµ, the longest wave length at which the bleaching rate was fast enough to study, the sum of the quantum energy and of the activation energy calculated from the temperature coefficients remains between 44 and 48.5 kg. cal. This result shows that in red light the energy deficit of the quanta can be made up by a contribution of thermal energy from the internal degrees of freedom of the rhodopsin molecule. The absorption spectrum of rhodopsin, which is not markedly temperature-dependent at shorter wave lengths, also becomes temperature-dependent in red light of wave lengths longer than about 570 to 590 mµ. The temperature dependence of the bleaching rate is at least partly accounted for by the temperature coefficient of absorption. There is some evidence that the temperature coefficient of bleaching is somewhat greater than the temperature coefficient of absorption at wave lengths longer than 590 mmicro;. This means that the thermal energy of the molecule is a more critical factor in bleaching than in absorption. It shows that some of the molecules which absorb energy-deficient quanta of red light are unable to supply the thermal component of the activation energy needed for bleaching, so bringing about a fall in the quantum efficiency. The experiments show that there is a gradual transition between the activation of rhodopsin by light and the activation by internal energy. It is suggested that energy can move freely between the prosthetic group and the protein moiety of the molecule. In this way a part of the large amount of energy in the internal degrees of freedom of rhodopsin could become available to assist in thermal activation. Assuming that the minimum energy required for bleaching is 48.5 kg. cal., an equation familiar in the study of unimolecular reaction has been used to estimate the number of internal degrees of freedom, n, involved in supplying the thermal component of the activation energy when rhodopsin is bleached in red light. It was found that n increases from 2 at 590 mµ to a minimum value of 15 at 750 mµ. One wonders what value n has at 1050 mµ, where vision still persists, and where rhodopsin molecules may supply some 16 kg. cal. of thermal energy per mole in order to make up for the energy deficit of the quanta.     

PHOTOLYTIC LIPIDS FROM VISUAL PIGMENTS

A method is described for the preservation of iodopsin, the labile photopigment of daylight vision, by freeze drying in vacuo. The lipids released by the action of light on rhodopsin and iodopsin are found to be similar and to possess a labile absorption spectrum in chloroform, with a rising peak at about 390 mµ and a declining peak in the region of 470 mµ. After the change is complete the absorption spectrum resembles closely that of retinene.     

THE QUANTUM YIELD OF HYDROGEN AND CARBON DIOXIDE ASSIMILATION IN PURPLE BACTERIA

1. The effect of H₂ tension, CO₂ tension, pH, time, light intensity, density of suspension, salt content of the medium, and certain spectral regions on the rate of photoassimilation of H₂ and CO₂ by Streptococcus varians has been studied. 2. The method of making light absorption measurements with thin suspensions of bacteria is described. 3. A light source, optical system, and filter for isolating 852 mµ with 894 mµ in sufficient intensity for photochemical work and an improved design of thermostat are given. 4. The photoassimilation of 2H₂ with 1CO₂ apparently involves little over all energy change but nevertheless requires 4 quanta.     

Spectral Sensitivity of the Common Prawn, Palaemonetes vulgaris

The vision of Palaemonetes is of particular interest in view of extensive studies of the responses of its chromatophore systems and eye pigments to light. The spectral sensitivity is here examined under conditions of dark adaptation and adaptation to bright colored lights. In each case the relative number of photons per one-fiftieth sec flash needed to evoke a constant peak amplitude (usually 25 or 50 µv) in the electroretinogram (ERG) was measured at various wavelengths throughout the spectrum. The sensitivity is the reciprocal of this number. In dark-adapted animals the spectral sensitivity curve consists of a broad, almost symmetrical band, maximal at about 540 mµ, with a shoulder near 390 mµ. Adaptation to bright red or blue light, left on continuously throughout the measurements, depresses the 540 mµ peak without notably changing its shape or position, implying that only one visual pigment operates in this region. Adaptation to red light, however, spares a violet-sensitive system, so that a high, narrow peak at 390 mµ now dominates the spectral sensitivity function. The 540 and 390 mµ peaks are apparently associated with different visual pigments; and these seem to be segregated in different receptor systems, since the associated ERG's have markedly different time constants. It is suggested that these two sensitivity bands may represent the red- and violet-sensitive components of an apparatus for color differentiation.     

THE EFFECTIVENESS OF THE SPECTRUM IN CHLOROPHYLL FORMATION

1. Although the carotenoid pigments are present in large concentration in the plastids of etiolated Avena seedlings as compared with protochlorophyll, the pigment precursor of chlorophyll, it is possible to show that the carotenoids do not act as filters of the light incident on the plant in the blue region of the spectrum where they absorb heavily. This suggests that the carotenoids are located behind the protochlorophyll molecules in the plastids. 2. Since the carotenoids do not screen and light is necessary for chlorophyll formation, an effectiveness spectrum of protochlorophyll can be obtained which is the reciprocal of the light energy necessary to produce a constant amount of chlorophyll with different wavelengths. The relative effectiveness of sixteen spectral regions in forming chlorophyll was determined. 3. From the effectiveness spectrum, one can conclude that protochlorophyll is a blue-green pigment with major peaks of absorption at 445 mµ, and 645 mµ, and with smaller peaks at 575 and 545 mµ. The blue peak is sharp, narrow, and high, the red peak being broader and shorter. This differs from previous findings where the use of rougher methods indicated that red light was more effective than blue and did not give the position of the peaks of absorption or their relative heights. 4. The protochlorophyll curve is similar to but not identical with chlorophyll. The ratio of the peaks of absorption in the blue as compared to the red is very similar to chlorophyll a, but the position of the peaks resembles chlorophyll b. 5. There is an excellent correspondence between the absorption properties of this "active" protochlorophyll and what is known of the absorption of a chemically known pigment studied in impure extracts of seed coats of the Cucurbitaceae. Conclusive proof of the identity of the two substances awaits chemical purification, but the evidence here favors the view that the pumpkin seed substance, which is chemically chlorophyll a minus two hydrogens, is identical with the precursor of chlorophyll formation found in etiolated plants.     

The Components of the Visual System of a Dragonfly

In this study of the electroretinograms of dragonflies (adults and nymphs) the objectives were to determine the number of classes of photoreceptors present in the visual system and to allocate these to particular morphological regions. There are probably five classes of photoreceptors present with peak sensitivities near 550, 530, 518, 420, and < 380 mµ. The dorsal ocelli contain two classes (518 mµ and < 380 mµ). The ventral (anterior) ommatidia of the adult compound eye contain at least two classes (near 518 mµ and < 380 mµ) and probably a third class (near 550 mµ). The dorsal ommatidia of the adult compound eye contain one class (420 mµ) and possibly another class (< 380 mµ). The compound eye of the nymph contains one class (530 mµ) and possibly another class (420 mµ).     

Laser activation of rapid absorption changes in spinach chloroplasts and chlorella

The kinetics of the 520 mμ absorption change in spinach chloroplasts and Chlorella vulgaris following a flash from the ruby laser have been determined as follows: rise halftime ≤ 0.3 × 10⁻⁶ second; rapid recovery halftime = 5 to 6 × 10⁻⁶ second; intermediate recovery halftime = 4 × 10⁻⁴ second (spinach chloroplasts only); slow recovery halftime = 12 to 170 × 10⁻³ second, dependent on the measuring light intensity and aerobicity of the suspension.

The rapid phase of the 520 mμ reaction is approximately independent of temperature, from 295° to 77° Absolute.

With increasing oxygenation of the sample, the extent of the rapid phase decreases, the extent of the slow phase increases, while the extent of the intermediate phase in spinach chloroplasts remains constant.

In spinach chloroplasts, no recovery halftime of the 3 recovery phases for the 520 mμ absorption change was observed to correspond to the halftime for oxidation of cytochrome f (t_½ = 1.3 × 10⁻³ second).

     

Effect of temperature on electrolyte metabolism of isolated frog skin

A study is presented on the effect of temperature on unidirectional active ion transport, resting electrolyte equilibrium (electrolyte composition), and oxygen consumption in isolated frog skin. The aims were twofold: first, to find out whether the rate of active transport can be changed without affecting the Na⁺ and K⁺ balance of skin itself; second, to arrive at minimal ΔNa/ΔO₂ values by correlating quantitatively inhibition of active ion transport with inhibition of O₂ consumption. NaCl transport was maximal at 20°C. At 28° and at temperatures below 20°, rate of NaCl transport was diminished. In many instances NaCl transport was diminished in skins which maintained their normal Na⁺ and K⁺ content. In several cases, however, neither rate of transport nor resting electrolyte equilibrium was affected; in other cases, both were. O₂ consumption decreased when lowering the temperature over the range from 28 to 10°C. From a plot of log Q_OO2 against 1/T an activation energy of µ 13,700 cal. was calculated, valid for the range from 10 to 20°C. It appeared that µ was smaller for temperatures above 20°C. Working between 10 and 20°, it was found that, on the average, 4 to 5 equivalents of Na⁺ were transported for one mole of O₂ consumed in skins with undisturbed resting electrolyte equilibrium.     

Action spectra for photoreactivation of ultraviolet-irradiated Escherichia coli and Streptomyces griseus

Action spectra for photoreactivation (light-induced recovery from ultraviolet radiation injury) of Escherichia coli B/r and Streptomyces griseus ATCC 3326 were determined. The spectral region explored was 365 to 700 mµ. The action spectrum for S. griseus differed from that for E. coli, indicating that the chromophores absorbing reactivating energy in the two species were not the same. Reactivation of S. griseus occurred in the region 365 mµ (the shortest wave length studied) to about 500 mµ, with the most effective wave length lying near 436 mµ. This single sharp peak in the spectrum at 436 mµ suggested the Soret band typical of porphyrins. Reactivation of E. coli occurred in the region 365 to about 470 mµ, with the most active wave length lying near 375 mµ. The single, non-pronounced peak near 375 was probably not due to a Soret band, and the identification of the substance absorbing reactivating light in E. coli is uncertain. In neither species was the region 500 to 700 mµ active. The implications of these action spectra and their differences are discussed.     

The Spectral Sensitivity of Crayfish and Lobster Vision

(1) The spectral sensitivity function for the compound eye of the crayfish has been determined by recording the retinal action potentials elicited by monochromatic stimuli. Its peak lies at approximately 570 mµ. (2) Similar measurements made on lobster eyes yield functions with maxima in the region of 520 to 525 mµ, which agree well with the absorption spectrum of lobster rhodopsin if minor allowances are made for distortion by known screening pigments. (3) The crayfish sensitivity function, since it is unaffected by selective monochromatic light adaptation, must be determined by a single photosensitive pigment. The absorption maximum of this pigment may be inferred with reasonable accuracy from the sensitivity data. (4) The visual pigment of the crayfish thus has its maximum absorption displaced by 50 to 60 mµ towards the red end of the spectrum from that of the lobster and other marine crustacea. This shift parallels that found in both rod and cone pigments between fresh water and marine vertebrates. In the crayfish, however, an altered protein is responsible for the shift and not a new carotenoid chromophore as in the vertebrates. (5) The existence of this situation in a new group of animals (with photoreceptors which have been evolved independently from those of vertebrates) strengthens the view that there may be strong selection for long wavelength visual sensitivity in fresh water.