AND LIFE-HISTORY OF A TROPICAL EPIPHYLLOUS LICHEN. 95 
regularly, there is no reason why the above disk-like structure and form should be 
altered. Аз a matter of fact, however, a perfectly regular circle is not maintained for 
long: at certain points in their progress on the surface of the leaf, the marginal cells 
meet with minute obstructions to their spread, while individual cells of the thallus 
produce hairs &c. as described, and others cease to divide in the usual manner. At any 
such point of arrested or modified growth * an indentation of the margin is apt to be 
produced as the surrounding series of cells grow forth into a lobe or branch. In some 
eases these lobes and sulci are developed with such regularity that a more or less stellate 
form is assumed by the older thallus (Pl. XVIII. fig. 2); at other times, and especially 
where the attacks of a fungus &c. or some diseased conditions intervene, very irregular, 
one-sided, and apparently lawless growths are set up. 
Nevertheless even a single lobe of an advanced thallus suffices to illustrate the manner 
of growth and sequence of cell-divisions described. In fig. 5с, for instance, and іп 
Pl. XX. figs. 39 & 40, this is fully attested; and even in cases where the numerous 
checks encountered in the horizontal extension of an advanced thallus have caused so 
many aberrations that all regularity appears at first sight to have been destroyed, inves- 
tigation of the separate portions confirms the truth of the above remarks. 
When the disk has thus increased in size, and produced a thallus of which certain cells 
give off rhizoids, others, marginal and subüerial, barren and fertile hairs, as described, it 
usually happens that certain older cells become transformed into what are presumably 
some kind of reproductive organs, as follows :—The cell becomes swollen, with increase in 
volume of its fat-and-starch-containing protoplasm, into an ovoid sac, the walls of which 
bulge on all sides, pushing apart those in its immediate neighbourhood as they do so 
(figs. 39 & 40 &c.) The dense orange-red protoplasmic contents then become slowly 
rearranged into finely granular masses, and break up into a number of spheroids tightly 
packed in the swollen sac, and each of which finally separates as a distinct naked mass of 
protoplasm. At this stage the contents give a deep blue reaction with iodine; and 
alcohol causes the spherical bodies to contract and appear like small shot tightly com_ 
pressed in a case (figs. 19@ & 42). Ata later date these escape as zoospore-like motile 
bodies, each moving by means of two long flagella placed at the anterior end (Pl. XX. 
& ХХІ. figs. 43, 45-47). | 
If Pl. XIX. fig. 16 and Pl. XX. fig. 39 be compared, it will be seen that these sacs 
containing zoospores arise at definite points in the thallus, and that each is the terminal 
cell of a given radial row of thallus-cells. Іп fig. 40 а have been depicted two of these 
transformed cells, with the adjacent vegetative cells in outline; and at 0 is given a semi- 
diagrammatie figure, whieh would be produced if the adjacent radial cell-rows were sepa- 
rated, as sometimes occurs in fact (fig. 6 4). Two events come out distinctly in this 
typical figure :—first, that the ovoid sac is the terminal cell of a series, as said; and, 
secondly, that the regularity of the typically dichotomous mode of branching becomes 
disturbed by this suppression of growth of one arm of the cell-series. In fig. 12 the cha- 
racter of the ovoid sac and its relation tothe rest of the thallus is made out with equal 
* The marginal short hairs are frequently found in tufts; and these conclude the extension of several con- 
tiguous cells. 
