2 4 6 8 10 12 14 16 Cells per colony

2 4 6 8 10 12 14 16 Cells per colony which however is lost if space between cells is progressively plugged by more, dense cells that add nothing to the hydrodynamic resistance.

It is a satisfying piece of teleology that the maximum advantage should seem to be achieved at the most typical coenobial size. Indeed, there is further observational and experimental evidence that the shape determines more of the sinking behaviour of stellate colonies than any of the Stokesian components. First, the argument would need to hold for Tabellaria flocculosa var., whose robust cells form eight-armed stellate colonies that contrast with the more typical habits adopted by this genus (see Knudsen, 1953). Frequently, up to 16 cells comprise the colonies but they remain in adherent pairs, preserving a regular eight-radiate form.

It is also of significance that some of the entries in Fig. 2.13 were derived from another study of Asterionella, using different source material, but its results fitted comfortably into the pattern. As pointed out earlier, several separate studies of the sinking rates of typical, eight-celled colonies of Asterionella, using quite disparate sources of algae, having differing dimensions and, possibly, densities, nevertheless achieved strikingly similar results (generally ws = 5.5-11.0 /.xm s-1), provided that the material used was moribund, poisoned or killed prior to measurement. However, the clearest demonstration that the sinking characteristics are strongly governed by shape comes from the somewhat fortuitous experimentation opportunity presented to Jaworski et al. (1988). When diatoms grow and divide, the replication of the siliceous cell wall is achieved by the development of a 'new' valve inside each of the two mother valves. When the daughters are eventually differentiated as new, independent entities, one has the dimensions of its parent cell but the other is slightly smaller, being no larger than the smaller of the parental valves (Volcani, 1981; Li and Volcani, 1984; Crawford and Schmid, 1986). In a clone of successive generations, one cell retains the parental dimensions but all the others are, to varying extents, smaller. Average size must diminish in proportion to the number of divisions. This is clearly not a process without limit, as the species-specific sizes of diatoms normally remain within stable and predictable limits. Size is recovered periodically through distinctive auxospore stages, which give rise to vegetative cells, with relatively large overall dimensions and large sap vacuoles. This process is observed in culture as well as in nature. When it was noticed, however, that sub-cultured cells of Asterionella clone L354, one of those isolated from wild types and maintained at the Ferry House Laboratory of the Freshwater Biological Association, were becoming unusually small, it was decided to include it in the sinking studies that were then being conducted in the laboratory. These observations were commenced in 1981 and were continued until 1986 (Jaworski et al., 1988). The clone never did recover size but the rate of growth became extremely slow towards the end. By the time of the last measurements, late in 1985, the cells had shrunk from long cuboids, 65 /j.m in length, to stubby ovoids,

1 Plot of sinking rates (ws) of killed stellate colonies of Asterionella formosa of comprising varying numbers of cells, compared with the sinking rates calculated for a spheres of the same volume and density (ws calc). The ratio, 0r = (ws calc)/ws is also shown against the horizontal axis. Redrawn from Reynolds (1984a).

0 0

Post a comment