trumpet-hyphae. In Nereocystis and Macrocystis a zone of tubes occurs, which present the appearance of sieve-tubes even to the eventual obliteration of the perforations by a callus. While there is a general tendency in the group to mucilaginous degeneration of the cell-wall, in Laminaria digitata there are also glands secreting a plentiful mucilage. Secondary growth in thickness is effected by the tangential division of superficial cells. The most fundamental external differentiation is into holdfast and shoot. In Laminariaceae secondary cylindrical props arise obliquely from the base of the thallus. In epiphytic forms the rhizoids of the epiphyte often penetrate into the tissue of the host, and certain epiphytes are not known to occur excepting in connexion with a certain host; but to what extent, if any, there is a partial parasitism in these cases has not been ascertained. In filamentous forms there is a differentiation into branches of limited and branches of unlimited growth (Sphacelaria). In Laminariaceae there is a distinction of stipe and blade. The blade is centrally-ribbed in Alaria and laterally-ribbed in Macrocystis. It is among the Sargassaceae that the greatest amount of external differentiation, rivalling that of the higher leafy plants, is reached. A characteristic feature of the more massive species is the occurrence of air-vesicles in their tissues. In Fucus vesiculosus they arise in lateral pairs; in Ascophyllum they are single and median; in Macrocystis one vesicle arises at the base of each thallus segment; in Sargassum and Halidrys the vesicles arise on special branches. They serve to buoy up the plant when attached to the sea-bottom, and thus light is admitted into the forest-like growths of the gregarious species. When such plants are detached they are enabled to float for great distances, and the great Sargasso Sea of the North Atlantic Ocean is probably only renewed by the constant addition of plants detached from the shores of the Caribbean Sea and Gulf of Mexico.
Growth in length is effected in a variety of ways. In Dictyota, Sphacelariaceae and Fucaceae there is a definite apical cell. In the first it is a biconvex lens, from which segments are continually cut off parallel to the posterior surface; and in the second an elongated dome, from which segments are cut off by a transverse wall. While, however, in Dictyota the product of the subsequent division in the segment enlarges with each subdivision, the divisions in the cylindrical segment of Sphacelariaceae are such that the whole product after subdivision, however many cells it may consist of, does not exceed in bulk the segment as cut off from the apical cell. In Dictyotaceae the apical cell occasionally divides longitudinally, and thus the dichotomous branching is provided for. In some Sphacelariaceae branches may appear at their inception as lateral protuberances of the apical cell itself. In Fucaceae an apical cell is situate at the surface of the thallus in a slit-like depression at the apex. From this cell segments are cut off in three or four lateral oblique planes.
A peculiar manner of growth in length is that to which the term trichothallic has been applied. It may readily be observed that in the hair-like branches of Ectocarpaceae, the point at which most rapid division occurs is situate near the base of the hair. In Desmarestia and Arthrocladia, for example, it is found that the thallus ends in a tuft of such hairs, each of them growing by means of an intercalated growing point. In these cases, however, the portions of the hairs behind the growing region become agglutinated together into a solid cylindrical pseudo-parenchymatous axis. In Cutleria the laminated thallus is formed in the same way. The intercalated growing region of Laminaria affords an example of another variety of growth in Phaeophyceae. While the laminated portion of the thallus is being gradually worn off in our latitudes during the autumnal storms, a vigorous new growth appears at the junction of the stipe and the blade, as the result of which a new piece is added to the stipe and the lamina entirely renovated.
Both asexual and sexual reproduction occur among Euphaeophyceae. Fucaceae are marked by an entire absence of the asexual method. The sexual organs—oogonia and antheridia—are borne on special portions of the thallus in cavities known as conceptacles. Both organs may occur in one conceptacle, as in Pelvetia, or each may be confined to one conceptacle or even one plant, as in Fucus vesiculosus. The oogonia arise on a stalk cell from the lining layer of the cavity, the contents dividing to form eight oospheres as in Fucus, four as in Ascophyllum, two as in Pelvetia, or one only as in Halidrys. It would seem that eight nuclei primarily arise in all Fucaceae, and that a number corresponding to the number of oospheres subsequently formed is reserved, the rest being discharged to the periphery, where they may be detected at a late stage. On the maturation of the oospheres the outer layer of the oogonial wall ruptures, and the oospheres, still surrounded by a middle and inner layer, pass out through the mouth of the conceptacle. Then usually these layers successively give way, and the spherical naked oospheres float free in the water. The antheridia, which arise in the conceptacular cavity as special cells of branched filaments, are similarly discharged whole, the antherozoids only escaping when the antheridia are clear of the conceptacle. The antherozoids are attracted to the oospheres, round each of which they swarm in great numbers. Suddenly the attraction ceases, and the oosphere is fertilized, probably at that moment, by the entry of a single antherozoid into the substance of the oosphere; a cell-wall is formed thereupon, in some cases in so short an interval as five minutes. Remarkable changes of size and outline of the oosphere have recently been described as accompanying fertilization in Halidrys. Probably the act of fertilization in plants has nowhere been observed in such detail as in Fucaceae. Dictyotaceae resemble Fucaceae in their pronounced oogamy. They differ, however, in being also asexually reproduced. The asexual cells are immotile spores arising in fours in sporangia from superficial cells of the thallus. In Dictyota the oospheres arise singly in oogonia, crowded together in sori on the surface of the female plant. The antheridia have a similar origin and grouping on the male plant. Until the recent discovery by Williams of motility, by means of a single cilium, of the antherozoids of Dictyota and Taonia, they were believed to be immotile bodies, like the male cells of red seaweeds. In Dictyota the unfertilized oosphere is found to be capable of undergoing a limited number of divisions, but the body thus formed appears to atrophy sooner or later.
Of the small family of the Tilopteridaceae our knowledge is as yet inadequate, but they probably present the only case of pronounced oogamy among Phaeosporeae. They are filamentous forms, exhibiting, however, a tendency to division in more than one plane, even in the vegetative parts. The discovery by Brebner of the specific identity of Haplospora globosa and Scaphospora speciosa marks an important step in the advance of our knowledge of the group. Three kinds of reproductive organs are known: first, sporangia, which each give rise to a single tetra-, or multi-nucleate non-motile, probably asexual spore; second, plurilocular sporangia, which are probably antheridia, generating antherozoids; and third, sporangia, which are probably oogonia, giving rise to single uninucleate non-motile oospheres. No process of fertilization has as yet been observed.
The Cutleriaceae exhibit a heterogamy in which the female sexual cell is not highly specialized, as it is in the groups already described. From each locule of a plurilocular sporangium there is set free an oosphere, which, being furnished with a pair of cilia, swarms for a time. In similar organs on separate plants the much smaller antherozoids arise. Fertilization has been observed at Naples; but it apparently depends on climatic conditions, as at Plymouth the oospheres have been observed to germinate parthenogenetically. The asexual organs in the case of Cutleria multifida arise on a crustaceous form, Aglaozonia reptans, formerly considered to be a distinct species. They are unilocular, each producing a small number of zoospores.
The possession of two kinds of reproductive organs, unilocular and plurilocular sporangia, is general among the rest of the Phaeosporeae. Bornet, however, called attention in 1871 to the fact that two kinds of plurilocular sporangia occurred in certain species of the genus Ectocarpus—somewhat transparent organs of an orange tint producing small zoospores, and also more opaque organs of a darker colour producing relatively larger zoospores. On the discovery of another such species by F. H. Buffham, Batters in 1892 separated the three species, Ectocarpus secundus, E. fenestratus, E. Lebelii, together with the new species, into a genus, Giffordia, characterized by the possession of two kinds of plurilocular sporangia. The suspicion that a distinction of sex accompanied this difference of structure has been justified by the discovery by Sauvageau of undoubted fertilization in Giffordia secunda and G. fenestrata. The conjugation of similar gametes, arising from distinct plurilocular sporangia, was observed by Berthold in Ectocarpus siliculosus and Scytosiphon lomentarius in 1880; and these observations have been recently confirmed in the case of the former species by Sauvageau, and in the case of the latter by Kuckuck. In these cases, however, the potential gametes may, failing conjugation, germinate directly, like the zoospores derived from unilocular sporangia. The assertion of Areschoug that conjugation occurs among zoospores derived from unilocular sporangia, in the case of Dictyosiphon hippuroides, is no doubt to be ascribed to error of observation. It would thus seem that the explanation of the existence of two kinds of sporangia, unilocular and plurilocular, among Phaeosporeae, lies in the fact that unilocular sporangia are for asexual reproduction, and that plurilocular sporangia are gametangia—potential or actual. It must, however, be remembered that so important a generalization is as yet supported upon a somewhat narrow base of observation. Moreover, for the important family of the Laminariaceae only unilocular sporangia are known to occur; and for many species of other families, only one or other kind, and in some cases neither kind, has hitherto been observed. The four species—Ectocarpus siliculosus, Giffordia secunda, Cutleria multifida and Haplospora globosa—may be taken to represent, within the Phaeosporeae, successive steps in the advance from isogamy to oogamy.
The Peridiniaceae have been included among Flagellata under the title of Dinoflagellata. The majority of the species belong to the sea, but many are found in fresh water. The thallus is somewhat spherical and unicellular, exhibiting a distinction between anterior and posterior extremities, and dorsal and ventral surfaces. The wall consists of a basis of cellulose, and in some cases readily breaks up into a definite number of plates, fitting into one another like the plates of the carapace of a tortoise; it is, moreover, often finely sculptured or coarsely ridged and flanged. Two grooves are a constant feature of the family, one running transversely and another longitudinally. In these grooves lie two cilia, attached at the point of meeting on the dorsal surface. The protoplast is uninucleate and vacuolate, and contains chromatophores of a brownish colour. It is not clear that
