Reynolds ( 1984a)
a Methods: (1) calculations based on experimental changes in velocity and gas-vesicle content (see text); (2) deteminations by centrifugation through artificial density gradient to isopycny;
(3) gravimetric determinations of mass and volume.
b Range covers colonies or filaments with maximum known gas vacuolation to colonies or filaments after subjection to pressure collapse (see text).
c Post-auxosporal (wide filaments).
d Pre-auxosporal (narrow filaments).
All measurements on wild material, unless otherwise stated.
but which are heavier than fresh water if the vesicles are collapsed by pressure treatment (see below). Two, small unarmoured chlorophytes are included (data of Oliver et al., 1981); how typical they are of non-siliceous algae is not known. The silica-clad diatoms have densities generally >1100 kgm-3, although there is a good deal of variability. Interestingly, it seems likely that density varies inversely to internal volume (Asterionella vs. Stephanodiscus, slender pre-auxosporal Aulacoseira vs wide post-auxosporal cells).
Apart from these generalisations, average densities of many planktic algae are plainly influenced by a number of discrete mechanisms. These include lipid accumulation, ionic regulation, mucilage production and, in Cyanobacteria, the regulation of gas-filled space.
Fats and oils normally account for some 2-20% of the ash-free dry mass of phytoplankton cells, perhaps increasing to 40% in some instances of cellular senescence (Smayda, 1970; Fogg and Thake, 1987). Most lipids are lighter than water and, inevitably, their presence counters the normal excess density to some limited extent. Oil accumulation is responsible for the ability of colonies of the green alga Botryococcus to float to the surface in small lakes and at certain times of population senescence (Belcher, 1968). However, it is improbable that oil or lipid storage could reverse the tendency of diatoms to sink. Reynolds (1984a) calculated that were the entire internal volume of an Asterionella cell to be completely filled with the lightest known oil, its overall density (~1005 kgm-3) would still not be enough to make it float. Walsby and Reynolds (1980) concluded that the reduction in density consequential upon intracellular lipid accumulation would, unquestionably, contribute to a reduced rate of sinking but they doubted any primary adaptive significance, neither was there evidence of its use as a buoyancy-regulating mechanism.
Inherent differences in the densities of equimo-lar solutions of organic ions raise the possibility that selective retention of 'light' ions at the expense of heavier ones could enable organisms to lower their overall densities. In a classical paper, Gross and Zeuthen (1948) calculated the density of the cell sap of the marine diatom Dity-lum brightwellii to be ~1020 kgm-3, that is, significantly lower than the density of the suspending sea water and actually sufficient to bring overall density of the live diatom close to neutral buoyancy. The density difference between sap and sea water was explicable on the basis of a substantial replacement of the divalent ions (Ca2+, Mg2+) by monovalent ones (Na+, K+) with respect to their concentrations in sea water. Some years later, Anderson and Sweeney (1978) were able to follow changes in the ionic composition of cell sap of Ditylum cells grown under alternating light-dark periods. They were able to show that density may, indeed, be varied by up to ±15 kgm-3, through the selective accumulation of sodium or potassium ions, though interestingly, not sufficiently to overcome the net negative buoyancy of the cells. Elsewhere, Kahn and Swift (1978) were investigating the relevance of ionic regulation to the buoyancy of the dinoflagellate Pyrocys-tis noctiluca; they showed that by selective adjustment of the content of Ca2+, Mg2+ and (SO4)2-, the alga could become positively buoyant.
The effectiveness of this mechanism is not to be doubted but its generality must be regarded with caution. For it to be effective does depend upon maintaining a relatively large sap volume. The scope of density reduction is limited, in so far as the dominant cations in sea water are the lighter ones and the lowest sap density is the isotonic solution of the lightest ions available. The scope for ionic regulation in phytoplankton of freshwater (pw < 1002 kgm-3) is too narrow to be advantageously exploited.
The mucilaginous investment that is such a common feature of phytoplankton, especially of freshwater Cyanobacteria, chlorophytes and chrys-ophytes, has long been supposed to function as a buoyancy aid. Again, that mucilage does reduce overall density is, generally, indisputable. Whether this is a primary function is less certain (see Box 6.1, p. 271) and it is mathematically demonstrable that the presence of a mucilaginous sheath does not always reduce sinking rate; in fact it may positively enhance it.
The presence, relative abundance and consistency of mucilage is highly variable among phy-toplankton. Mucilages are gels formed of loose networks of hydrophilic polysaccharides which, though of high density (~1500kgm-3) themselves, are able to hold such large volumes of water that their average density (pm) approaches isopycny. Reynolds et al. (1981) estimated the density of the mucilage of Microcystis to average pw + 0.7 kgm-3. The presence of mucilage cannot make the organism less dense than the suspending water but it can bring the the average density of the cell or colony maintaining it much closer to that of the medium (pc). However, the clear advantage that this might bring to reducing ws in, for instance, Eq. (2.16) must be set against a compensatory increase in overall size (ds is increased). Thus, for mucilage to be effective in depressing sinking rate, the density advantage must outweigh the disadvantage of increased size.
This relationship was investigated in detail by Hutchinson (1967). If a spherical cell of density, pc, is enclosed in mucilage of density, pm, such that its overall diameter is increased by a factor a, then its sinking rate will be less than that of the uninvested cell, provided:
Because a is always >1, the density difference between cell and mucilage must be at least that between mucilage and water. Supposing pc to be ~1016kgm-3, pw to be 999 and pm to be 999.7,
Eq. (2.17) can be solved as 23.3 > a (a + 1), or that a must be <4.3, if the presence of mucilage is to reduce the sinking rate. The maximum advantage can be solved graphically (see Fig. 2.9); with the nominated values, the greatest advantage occurs when a ~ 2.3. Of course, the precise optimal value of a varies from alga to alga, depending partly upon the nature of the mucilage but mainly upon the cell density. For a diatom with a cell density of 1200 kg m-3, the greatest value of a that would bring a net reduction in its sinking rate could be as high as 16.4, with maximum advantage at a ~ 8.7.
In order to compare the mucilage provision among planktic algae, which are not all spherical, it is useful first to express cell volume as a proportion of the total unit volume (vc/vc+m, as included in Fig. 2.9). Some examples are given in Table 2.4. In each instance, the range of values of a is calculated as the ratio of the diameter of the sphere equivalent to the full unit volume and the diameter of the sphere equivalent to the total volume of cells enclosed. The data presented appear to fall within the range of benefit (in terms of sinking rate), although in the case of Pseudosphae-rocystis and some Coenochloris colonies, the ratios appear to unfavourable, at least if the assumptions about the component densities apply in all these cases.
The surest way of lowering average density is to maintain gas-filled space within the protoplast. This is precisely what some of the plank-tic Cyanobacteria do and, as is well known, the organisms become buoyant at times, accumulatu-ing at the surface as a scum, constituting what was originally called a 'water bloom'. The term 'bloom' has since been applied to almost any planktic population (not even necessarily algal) significantly above the norm: it is another of those words that has been misused to the point of being rendered unhelpful. However, the biology of scum-forming Cyanobacteria is a fascinating topic, and not only because of the almost universal contempt in which most environmental and water-supply managers hold their unsightliness and potential toxicity (Bartram and Chorus, 1999; see also Section 8.3.2). Part of the remarkable account of the functional morphology and population dynamics concerns the ability of these Cyanobacteria to regulate the buoyancy provided by their gas vacuoles (Reynolds et al. 1987). Nested within this is the unfolding appreciation of the structure and function of the buoyancy provision itself. Much of the progress over the last 30 years has been spearheaded by A. E. Walsby and his coworkers. Walsby's (1994) review is one of the most comprehensive, and it is this to which the reader is referred for all details.
Here, it is sufficient to emphasise that, from the time their existence was first established (Klebahn, 1895), gas vacuoles have been assumed to have the function of providing buoyancy. Although this may have been neither their original nor their only function (Porter and Jost, 1973, 1976), these uniquely prokaryotic organelles certainly do reduce the average density of the cell in which they occur. They are not bubbles -surface tension is much too powerful to permit
3 The effect of mucilage thickness (a, as a multiple of diameter, ds) on the average density (p = pc + pm) of a spherical alga of constant diameter and density (pc = 1016 kg m-3). The arrow on the relative velocity plot indicates the point of maximum advantage of mucilage investment in the context of sinking-velocity reduction. Redrawn from Reynolds (1984a).
Table 2.4 Relative volumes of cell material (vc) as a proportion of the full unit volume (vc+m) of named mucilage-producing phytoplankters, expressed in terms of Hutchinson's (1967) factor a (see text)
Microcystis aeruginosa Anabaena circinalis
Chlamydocapsa planctonica Coenochloris fottii Eudorina unicocca
Pseudosphaerocystis lacustris Staurastrum brevispinum Fragilaria crotonensis vc/vc+m a
Reynolds et al. (1981) Previously unpublished measurements reported in Reynolds (1984a) Previously unpublished measurements reported in Reynolds (1984a) Previously unpublished measurements reported in Reynolds (1984a) Previously unpublished measurements reported in Reynolds (1984a) Previously unpublished measurements reported in Reynolds (1984a) From direct measurements taken from Fig. 27 of Ruttner (1953) From direct measurements of linear dimensions taken from Plates 1c and Id of Canter and Jaworski (1978), with calculation of a their existence at the scale of micrometres - but rigid stacks of proteinaceous cylindrical or prismatic envelopes called gas vesicles (Bowen and Jensen, 1965). In the Cyanobacteria, they generally measure between 200 and 800 nm in length. The diameters of isolated gas vesicles vary inter-specifically between 50 and 120 nm, but are reasonably constant within any given species. Each molecule of the specialised gas-vesicle protein has a hydrophobic end and they are aligned in ribs in the vesicle wall so that the entry of liquid water into the internal space is prevented. However, the vesicle wall is fully permeable to gases and in no sense do the vesicles hold gas under anything but ambient pressure. The gas inside the vesicles is usually dominated by nitrogen with certain metabolic by-products but it is clear that the gas composition is of much less significance than is the gas-filled space, which is created as the vesicle assembles. The structures are vulnerable to external pressure, including the internal turgor pressure of the cell. They have a certain strength but, once a critical pressure has been exceeded, they collapse by implosion. They cannot be reinflated; they can only be built de novo, although the gas-vacuole protein is believed to be recyclable.
The critical pressures of isolated gas vesicles are inversely correlated to their diameters (Hayes and Walsby, 1986). The higher is the critical pressure of the vesicles, the greater the hydrostatic pressure and, thus, the greater water depth they can withstand. Intriguingly, vesicle size, like organism size and shape, co-varies with the principal ecological ranges in which individual species occur and, arguably, the habitats to which they are best adapted. In Anabaena f los-aquae, a common scum-forming species in small eutrophic lakes, vesicles measuring about 85 nm in diameter have a critical pressure of 0.3 to 0.7 MPa (Dinsdale and Walsby, 1972). In Microcystis aeruginosa, a species sometimes found in larger and more physically variable lakes, vesicles averaging 70 nm in diameter have critical pressures in the order of 0.6-1.1 MPa (Reynolds, 1973b; Thomas and Walsby, 1985). The species of Plank-tothrix (formerly Oscillatoria) of deep, glaciated lakes in mountainous regions have vesicles measuring 60-65 nm that withstand pressures of between 0.7 and 1.2 MPa (Walsby and Klemer, 1974; Walsby et al, 1983; Utkilen et al, 1985a). Anabaena lemmermanni, a species more usually found distributed in the deep mixed layers of larger temperate lakes, also has much stronger vesicles than most other Anabaena species (0.93 MPa: Walsby et al., 1991). Vesicles from the oceanic Trichodesmium thiebautii were found to tolerate up to 3.7 MPa (Walsby, 1978).
It is now recognised that gas-vesicle size is subject to very strong selective pressure. Narrow gas vesicles are less efficient at providing buoyancy and, for a given yield of gas-filled space, they are assembled at greater energetic cost. Narrower ones should only be selected if the extra strength is required (Walsby and Bleything, 1988). Now that the genes controlling gas-vesicle assembly can be identified relatively easily (Beard etal., 1999, 2000), the selection by hydrographic events (for instance, incidences of deep mixing of Plank-tothrix populations) for the survival of relatively more of the stronger or relatively more of the weaker kind is one of the most elegant demonstrations of gene-based natural selection to have been contrived (Bright and Walsby, 1999; Davis et al., 2003).
The buoyancy-providing role of the gas vesicles has been studied for over 30 years. By preparing very thick suspensions of Cyanobacteria, placing them in specific-gravity bottles and then subjecting them to pressures sufficient to collapse the vesicles, the volume of gas displaced can be measured very accurately. Expressed relative to the cell volume, the percentage of gas-filled space is readily calculated (Walsby, 1971). Reynolds (1972) used this method to collect the data used to construct Fig. 2.10, which is included to show that there is, for any given alga, a poten-
Was this article helpful?