beautiful regularly-placed cells that are to be seen in Anadyomene. This genus resembles Udotea in the form of its filaments, but differs in the branches being separated by a cellular expansion of the frond, instead of being close side by side. In this respect it is intermediate between Udotea and Anadyomene.

1. C. Brownii, n. s.

HAB. Australia, R. Brown in Brit. Mus. A small fragment in my own collection from among Australian weeds. I have sent a portion of the latter to Dr. Harvey for his herbarium at Trinity College, Dublin. This is not the A. plicata of Agardh, described as having only a few cells of large size.

Professor Agardh describes another species, with doubt, under the name of A. obscura, thus :-" fronde cuneata, venis obsoletis, in mari australi ad insulam Graham; specimen dedit Gaudichaud. Radix subglobosa. Frons ex angustiori basi (quasi stipula) dilatata, cuneata, longitudine digitalis, unciam lata, sublobata; venæ uniplicatæ, sparsæ, obsolete rubræ. Color viridescens, luridus; substantia stipitis firmior, crassior, partis superioris membranacea."-C. A. Agardh, Species Alyarum, i. 400 (1823); Kütz. Spec. Algarum, p. 511.

This may be allied to Calomena.

Genus 2. ANADYOMENE.

The frond flabellate, stipitate, often imbricate at the base, formed of a succession of single ovate cells with minute cells in the interspaces; midrib trifid or radiately branched; the primary cell with a series of diverging cells at the tip like a fan, all or three or five of the largest of which bear at their tip a similar series of diverging cells and branches. The upper part of the side of the main cells with a series of small cells on each side placed at right angles with the main cell; the disk of the frond formed of numerous small cells; the margin of the frond formed of fan-like series of cells.

Anadyomene, Lamouroux, Pol. Flex. 365; Agardh, Spec. Algarum, 401; Kützing, Phyt. Gener. 254; Species Algarum, 511; Harvey, Nereis Bor. Am. iii. 49.

This genus appears to have a very extensive distribution; Wulfen and Lamouroux found it on the coasts of Europe, Webb and Berthelot at the Canaries, Professor Harvey in Australia, Gaudichaud in Rawak and the Sandwich Islands, La Sagra in Cuba, and Martius in the

Brazils. It is to be regretted that the specimens from these different localities have not been critically examined.

Dr. Harvey's character is excellent, viz. root fibrous; frond stipitate, membranaceous, leaf-like, flabellately veined; the veins confluent, radiating from the base to the margins pedately multifid, excessively branched, and everywhere closely anastomosing; fructification unknown.

"As Professor J. Agardh remarks (Alg. Medit. 24), it is related to Valonia, from which it differs chiefly in the lateral cohesion of the branches of the generating filament, and to which it bears the same relation that Codium does to Vaucheria. It is still more nearly related to Microdictyon, where the frond orms an open network." Harvey 1. c. 49.

Professor Harvey gives an interesting account of the development of the Florida specimens in his Nereis Boreali-Americana,' vol. iii. 49.

1. A. stellata; frond coriaceous, the cells ovate, narrow at the base, with several diverging cells at the tip, some of which elongate, and are proliferous at their apex; the upper part of the sides of the basal cell, with some large cells placed at right angles with the principal cells; the frond between the main fibres formed of numerous variously-sized cells. Lichenoides gelatinosum tenue reticulatum, Dillen. Musc. 138. t. 19. f. 21. Ulva stellata, Wulfen, Cr. Aquat. 6; Jacq. Collec. i. 321; Roth, Cat. Bot. ii. 243, 325. Anadyomene flabellata, Lamx. Pol. Flex. t. 11. f. 3; Bory, Nouv. Fl. Pélop. 78. t. 41. f. 5; Kützing, Sp. Alg. 511. A. stellata, C. A. Agardh, Sp. Alg. i. 400; Syst. 191; Mart. Fl. Bras. i. 25; Montag. in La Sagra, Cuba, 22; Webb and Berth. Fl. Canar. iv. 180.

HAB. Mediterranean, Wulfen, spec. in Brit. Mus. Coast of France, Lamouroux.

Var. Floridana; larger; midrib more branched; cells oblong, more ovate, not so narrow below.-A. flabellata, Harvey, Nereis BorealiAmericana, iii. 48. t. 44; excellent.

HAB. Florida Key West, Herb. Harvey and Gray.

I am by no means certain that the specimens from the coast of France, Florida, Cuba, and Brazil, combined in the above synonyma, are the same species, but I have not sufficient specimens at my command to determine the question.

I have only seen two small fragments of Wulfen's from the Mediterranean that were given to the Banksian collection by Dawson Turner,

and a series of specimens from Florida collected by Professor Harvey, which he most kindly presented to me.

If I could regard these Mediterranean specimens as fair types of the plant usually found there, I should decide that it was distinct from those from Florida. These small fronds only contain a very few large cells, very different in this respect from the Florida specimens, but, on the other hand, the specimen figured by Lamouroux, found in the "Mousse de Corse," more nearly resembles those from Florida, and one can hardly believe that the Corsican Alga he examined could have come from the coast of America.

2. "A. plicata; frond plicate; veins subtrichotomous."-C. A. Agardh, Sp. Alg. i. 400; Kützing, Sp. Alg. 511.

HAB. Island of Rawak, Gaudichaud.

"Differt a præcedente (4. stellata) statura minore, fronde maxime plicata, venis paucioribus trichotomis, cum in illa frons tota venis occupata est, hæc magis continua venis quibusdam membranam percurrentibus; habitus omnino Collematis."-C. A. Agardh, Species Algarum, i. 400, 1823.

3. A. Cutleria; frond membranaceous; the cells oblong, nearly as wide at the base, with several diverging cells at the upper part, each bearing a similar series of diverging cells at the apex; the frond between the main cells filled up with one or two series of large cells at right angles with their margin.

HAB. Bermuda.

Described from a fine specimen received by Miss Cutler from Bermuda, and presented by that lady to me with the rest of her exotic Alga. I have divided the specimen between the British Museum, Dr. Harvey, and my own collection.

4. A. Wrightii; frond imbricated, coriaceous; joints linear-elongate, several times longer than broad, with a radiating group of cylindrical branches at the tip, two to four of which are longer than the rest and proliferous at the tip; the branchlets near the margin five or six, shorter, radiating, of nearly equal length; the interspaces between the branches wide, and filled up with small subequal cells.-A. Wrightii, Harvey, mss.

HAB. Loochoo Islands, C. Wright, King's and Rogers's Exploring Expedition, 1853 and 1856.

Professor Harvey most kindly sent me this species to compare with

Anadyomene Brownii. Its study induces me to propose to divide the genus into two subgenera, thus :

:

1. The cells of main stem linear; interspaces between the main filaments and cells close on their sides, filled up with nearly equalsized minute cells.-STENOCYSTIS, for A. Wrightii.

2. The cells of main stem ovate; interspaces between the main filaments filled up with large very different-sized cells.-ANADYOMENE, for A. stellata and A. Cutleria.

Stenocystis is somewhat intermediate between Anadyomene and Calomena, but it evidently belongs to the genus to which I have referred it, as instead of the main filament being only forked, it is provided with radiating cells at the top.

Genus 3. GRAYEMMA.

Frond fan-shaped from a central root; the main stem and branches in the centre of the frond and lobes formed of three or four parallel close series of short cells in transverse bands.

This genus is very different in its structure from Anadyomene. In the latter, the series of cells that form the axis of the frond and its lobes is single, one cell on the end of the other like a Conferva, the end cell being crowned with a radiating group of cells.

In Grayemma the frond and its lobes are supported by a broad midrib, which is formed of several close parallel longitudinal series of cells, the cells on the side of the midrib giving off radiating groups of cells. The end of the midrib is branched, and is elongated by the development of a radiating group of cells at the end of the former one, and this is how the many series of cells in the midrib are formed, and why they look like what they really are, a continued succession of radiating groups of cells forming a thick midrib; the parietes of the cells are so thin that in the dry specimen the outer surface of the cell is sunk in leaving the side-margin elevated; from the side of the midrib arises a group of diverging cells, and on the apex of these are formed another series as the frond enlarges thus the branches on the midrib are gradually formed and lengthened.

:

The disk of the frond between the midribs is filled up with a very numerous series of cells much smaller in size and more numerous than in Anadyomene, consequently there is a much greater difference between VOL. IV. [FEBRUARY 1, 1866.]

E

the disk of the frond and the main stems than there is between the cells in Anadyomene, which is, as it were, all composed of numerous diverging cells only differing in size.

The cells on the upper part of the sides of the main series in Anudyomene are furnished with a series of rather large cells placed at right angles with them; there are only a very few very small cells so placed n Grayemma, and they are not to be seen except in a few places on the frond.

If the chain of cells of the two genera are compared, it will be found that in Anadyomene each cell gives off at the tip a radiating series of cells, some of which being larger than the rest form a branch which ať its apex again gives off a radiating group of cells, some of which are similarly elongated and are proliferous, so that the frond is composed of a succession of trifid and in some rare instances four- or more numerously-divided branches. In Grayemma, on the contrary, the series of cells remain unbranched as long as they are parallel, and after being parallel for a time some diverge to the left or to the right, and then form another stem, giving off diverging series of cells.

In Grayemma the midribs extend almost up to the edge of the frond with a single group of cells, forming a fan at the top quite close to the edge, which is very different from the structure seen in Anadyomene.

All the midribs and branches of the specimens I have been able to examine are formed of several parallel close series of cells, except the tips of some of the smaller branchlets, which consist of a series of two or three cells placed one on the other, and ending in cells diverging from the tip of the last one like a fan, except in two cases, one a slender branch, which starts from the midrib and extends to the margin; this branch consists of a single series of cells as in Anadyomene, about twice as long as they are broad; and only giving off a short single branch, not dividing into branchlets as in Anadyomene. The second example of a single series of cells occurs in a simple branch that runs parallel to the main stem, and at length becomes united to it, and then assumes a compound form. This branch can only be considered as a series of cells that has been accidentally diverted from its proper position in the growth of the plant, and assumes it again, but it shows that the main stems are composed of many single series of cells united into a bundle to form the thick midribs.