both Islands, yet Carpophyllum elongatum is not known south of the Bay of Islands, C. plumosum is chiefly a North Island species, whilst the Cystophoras are better developed in the South than in the North. Landsburgia is apparently most abundant in the neighbourhood of Cooks Straits, whilst a distinct species is known from the Chathams. These distinctions might be indefinitely multiplied, but so little is known of the content of many parts of the coast line and the special habitats of many species, that it is unsafe to venture very far on such generalizations.

I shall now endeavour first to follow out the argument of my earlier paper, and show how it has been modified by our increased knowledge, and then deal with the questions raised in a more general way. The increase in the number of recognised species may be shown briefly thus:

The list of British species is based on that of Holmes and Batters, which is the last complete list available. No doubt it could now be considerably extended. The one unmistakeable point of con-trast is the much greater richness of the New Zealand waters in Rhodophyceae. Varieties in both lists have been omitted, and similar groups only have been compared. Doubtless further investigations will show that New Zealand waters are as rich as those of Great Britain in Chlorophyceae: and possibly also in Phaeophyceae. There are, for example, over thirty-five Ectocarpaceae in Great Britain, while only four are recorded from New Zealand. These are imper-fectly known, and there are doubtless many more to be discovered.

It should also be noted that the two New Zealand lists—owing to certain alterations in the classification—are not strictly comparable. The transference of the Bangiales from the Chlorophyceae to the Proto-Florideae reduces the one group in the 1925 list and increases the other. In spite of this change the percentage of the Chlorophyceae has been increased in comparison with the Phaeophyceae; and now numbers fifty-two per cent. instead of the earlier forty-four per cent. However, little guidance can be obtained from these percentages, which will fluctuate largely with the nature of the investigations carried on, and which of course only correspond very roughly with the number of species actually present in the country. The compari-son with Great Britain—an area of equal extent in a similar latitude—shows that in these proportions we have a typically temperate flora. The high percentage of Florideae is doubtless due to the wide North and South range of our coasts; and not to any peculiar richness in the group. We have moreover an algal Flora averaging about three species to the genus. Hence we cannot in any way regard it as fragmentary or Oceanic in type, though certain small families of northern waters are missing from it. Figures for comparison may

be found in Murray, 1893, p. 70. Apart from geographic considera-tions, sea temperature is probably the chief controlling factor in the distribution of the seaweeds. The Florideae being generally shade plants, or more often submerged altogether, must depend more than the Chlorophyceae or the Phaeophyceae upon sea temperatures. Other seaweeds being exposed at low tide depend for their existence upon such additional factors, as surface temperatures and the humidity of the air, the nature of the coast, the salinity of the water, the amount of light received, and the times of neap and spring tides. There are other ecological factors at work in determining distribution, but it is not proposed to deal with these in this paper in considering the general external distribution.

In this respect, the first point to be noted is the amount of specific endemism. This is small compared with that of the flowering plants amounting only to about forty-one per cent., while high compared with that amongst the seaweeds in most other regions of the world, though it is difficult to obtain suitable figures for comparison with similar districts elsewhere. Great Britain, of course, is the most similar area, and there the endemism is almost negligible, as might be expected, from its situation in the midst of the land masses of the world. As Cockayne has pointed out (1921, p. 311) the endemism amongst the New Zealand Dicotyledon rises as high as eighty-five per cent; and though for obvious reasons the marine algae are com-paratively widely distributed, it is quite possible that in them the amount of endemism rises higher than elsewhere, except perhaps in Australia, where there is a remarkable number of endemic Floridean genera.

Yet, whilst' in the case of the flowering plants, increasing know-ledge has tended to increase the percentage of known endemism, here it has tended to decrease it. In 1894, I estimated it as fifty-two per cent. It now appears as forty-one per cent. This decrease is largely due to the increased number of species, now known to be common to New Zealand and Australia. Whereas I previously estimated our Australian element to be only fifteen per cent. of the total, it now appears to be thirty per cent. On the other hand, the percentage of what we would now term sub-Antarctic species has considerably decreased, though it is still a noteworthy feature of the list, amount-ing to perhaps seven, per cent. This includes a very few forms which are also found in the Antarctic proper, e. g., Ballia callitricha and B. hombroniana, but generally speaking there is no distinct element in New Zealand seaweeds which can be recognised as purely Antarctic. The widely distributed species amount to about sixteen per cent. in both lists. The remaining species (six per cent.) are those with wide but often discontinuous distribution, amongst which may perhaps be traced a small and scarcely noteworthy Polynesian element. We thus get the following results, which I now wish to consider, somewhat more in detail:

Now let us consider the endemic element. Dr. Cockayne (1. c. p. 311) defines five degrees of specific endemism; but it will be quite impossible to follow his arrangement here. However, it may be borne in mind, and we may at least distinguish those species, that show endemism of the first degree. These are those which belong to endemic genera, sub-genera, and distinct sections of genera. Now our endemic genera are twelve in number, ^* Ptilopogon, Herponema, Marginaria, Landsburgia, Craspedocarpus, Apophlaea, Abroteia, Streblocadia, Microcolax, Pleurostichidium, Pandorea, and Dacytly-menia. These may be regarded as a distinctively New Zealand element, involving eighteen species. Most of them require more investigation than they have received. Marginaria with two species is a most distinct fucaceous genus, whose reproduction has not yet been fully worked out, and a detailed account of it is much to be desired. Landsburgia though perhaps not so distinctive is in a similar position. It has also two species, one on the main islands, and one on the Chathams. Apophlaea again with two species, is a red seaweed of quite uncertain position. Pleurostichidium is a remarkable and little known Floridean parasitic on a brown, an unusual occurrence. Pandorea belongs to the Ceramiaceae, is monotypic and quite insuffi-ciently known. The other endemic genera are less aberrant, and more closely resemble forms known elsewhere, and will not be further considered here. They are usually monotypic. It may, however, be noted that none of these genera belong to the Chlorophyceae which are usually widely distributed, that eight are Floridean, and four Phaeophycean. Three, Herponema, Microcolax, and Pleurostichidium, are parasitic.

Associated with these endemic genera, are those which though not confined to New Zealand have a distinctively New Zealand section, or those whose species are chiefly confined to New Zealand. Amongst these may be mentioned Carpophyllum (which, however, is closely related to the widely distributed Sargassum), Cladhymenia, Euzoni-clla, and Pachymenia. The New Zealand species of these genera are probably not so distinctive as those in the first group. They would still, however, come under Cockayne's defintion of specific endemism in the first degree (Metamorphe and Aeodes until recently have been regarded as monotypic endemic genera, but the same or another species has in each case, recently been found in the North Pacific). This gives us another fifteen species to add to our previous eighteen, making in all a total of thirty-three, which show specific endemism of the first degree.

Here is perhaps the place to mention the algal genera confined to New Zealand and Australia. They are nineteen in number, viz., Perithalia, Notheia, Hormosira, Xiphophora, Cystophora, Gloiosi-phonia, Mychodea, Dicranema, Areschougia, Melanthalia, Curdiaea, Sarcocladia, Hymenocladia, Hemineura, Phitymophora, Aphanocladia, Lenormandia, Warrenia, Thamnoclonium. Several other genera are confined to New Zealand, Australia, and the Cape of Good Hope. It

is clear, therefore, that the general distribution of the southern element is, as might be expected, of an east and west character. In such genera, however, as Notheia, Hormosira, Xiphophora, Cystophora, Melanthalia, Curdiaea, Sarcocladia, and Hymenocladia, sufficient dis-tinctiveness is shown to constitute Australasia as a distinct algal region, and it should be so considered in the future.

The majority of the remaining New Zealand endemic species would no doubt appear under Cockayne's classification as showing endemism in the fourth degree. Many of them are found in the large, world-wide genera such as Nitophyllum, Gigartina, and Polysiphonia. It would serve no good purpose in the present state of our knowledge to attempt any further analysis of this form of specific endemism. However, something must be said about the distribution of the endemics through the sub-classes and families. In the Chlorophyccae there are only nine, and of these seven occur in the genus Cladophora. This number is quite likely to be reduced by further exploration in adjacent countries and islands. The cosmopolitan nature of the Ulva-ceae at any rate is well known, and possibly to be attributed to their great age, rather than to special means of distribution. No doubt, however, there are still a considerable number of the smaller Chloro-phyceae yet to be listed.

With the Phacophyceae it is different. Our brown sea weeds are almost exclusively Australian in their relationships. With the exccp-tion of the endemic species already dealt with, one or two sub-Antarctic species from the Aucklands, Lessonia variegata and several doubtful cases, all our known Phaeophyceae are also found in Aus-tralia, and, as we have already seen, many of the most remarkable. (e. g., Notheia, Hormosira, Xiphophora, Perithalia) belong to distinc-tively Australasian genera. Our Fucaceae, indeed, are all Austra-lasian if not endemic, with the exception of D'Urvillea, which though South Australian, is chiefly sub-Antarctic in its distribution. This is a remarkable fact, and I do not think the relationship can be attributed purely to drift across the Tasman sea in currents from the west driving before westerly winds, particularly when we remember that the Laminarians on the other hand, though more widely distributed, tend to show a sub-Antarctic facies and have their headquarters in Southern America. The Ectocarpaceae are more cosmopolitan, and are probably an older family. However, the New Zealand species are so little known, that there is little to be said about their distribution.

Not only are there a number of endemic Australasian genera, but of the thrity per cent. of our seaweeds that are also found in Australia, no less than one hundred and twenty-six, or twenty-six per cent. of the total New Zealand seaweeds, are confimed to New Zealand and Australia. About one thousand Australian seaweeds are known, therefore about thirteen per cent. of the Australian seaweeds are found only in New Zealand apart from Australia. The connection is obviously a very much closer one than it is for the Phanerogams; and it is not altogether represented by showing percentages. The numbers of individuals have to be considered. One may traversethe Australian forest for days and see few New Zealand species, and

those in small quantity, but if one goes to the sea-shore at Sydney or Melbourne the New Zealand species of seaweed are much in evidence. Possibly, but not probably, ocean currents are responsible for this close connection. In this case one might expect to find more Austra-lian species on the west than on the east coast of New Zealand; but much of the West-Coast is poor collecting ground, and in any case there has been but little examination of it. What is particularly required is a close examination of the species common to both Floras, and this has yet to be done. The Australian element of our algal flora contains, further, a distinct Malayan group, which, however, scarcely needs discussion.

The next element of especial interest is that now known as the sub-Antarctic. The term has been loosely and variously used by different writers. It is proposed here to consider the Antarctic as that region bounded by the northern limit of pack-ice; and the sub-Antarctic as bounded by the northern limit of drift-ice. Thus defined the Antarctic will include the Antarctic continent. The sub-Antarctic is entirely insular and includes Tierra del Fuego, the Falklands, South Georgia, Tristan da Cunha, the Crozets, Kerguelen, the Aucklands, and Campbell Island, with various other small islands. As is well known, these islands show striking resemblances in their animal and vegetable life, and for biological purposes form a well-marked region. When species from these groups pass northward into Tasmania, Southern Australia, New Zealand, and South Africa, they constitute a distinct element in the flora, which may be termed the Austral-sub-Antarctic. Now this Austral-sub-Antarctic group is in New Zealand represented by thirty-five species in the seaweeds, viz: Chaetomorpha Darwinii, Halopteris funicularis, Halopteris hordacea, Scytothamnus australis, Adenocystis utricularis, Lessonia variegata, D'Urvillea antarctica, Chaetangium variolosum, Gigartina fissa, Iridaea cordata, I. laminarioides, Callophyllis variegata, C. tenera, Rhodophyllis acanthocarpa, R. angustifrons, Rhodymenia corallina, Nitophyllum multinerve, N. Smithii, Phycordrys quercifolia, Schizoneura dichotoma, S. Dovisii, Ptilonia magellanica, Delisea pulchru, Chondria angustata, Lophurella comosa, L. Hookeriana, Poly-siphonia microcarpa, Herposiphonia ceratoclada, Griffithsia antarctica, Ballia callitricha, B. scoparia, Antithamnion flaccidum, A. ptilota, A. ternifolium, Lithothamnion antarcticum.

There is in addition a very small element, which is sub-Antarctic proper, reaching to the Aucklands and not penetrating to the main-land of New Zealand, e. g., Desmarestia Willii, Scytothamnus fascicu-latus. Now the main characteristic of these sub-Antarctic species is their wide and discontinuous distribution in south circumploar seas. I have given reasons elsewhere (Lg., 1895, p. 305) for doubting the frequency of living seaweeds crossing wide ranges of ocean. It is true that if cystocarpic specimens of red seaweed could cross the barrier, they might at times successfully reproduce themselves, and undoubtedly some of the larger brown seaweeds occasionally migrate from coast to coast; but the present distribution shows that such occurrences are rare; and for many species doubtless quite impossible. The currents of the Pacific are still imperfectly known; but the ice-

drift from the polar regions obviously indicates northward currents round the Antarctic circle, and that these sometimes carry seaweeds is well known. Thus, speaking of the Humboldt current between 150° W., and 117° W., Captain Crutchley (1891, p. 275) states: “This current cannot be overlooked even in summer, not because of the ice contained therein only, but also because of its peculiar colour, which, in two separate years, was a dirty ultramarine, and because it con-tained large quantities of seaweed and various kinds of floating matter.” Yet there are a number of seaweeds confined only to Ant-arctic seas (Skottsberg, 1907, s. 158), so it is clear that drifting weeds do not necessarily establish themselves on the shores on which they are cast. (v. also Kylin and Skottsberg, 1919, Rhodophyceen s. 80.) Whatever the explanation of the present distribution of our seaweeds may be, it is clear from this, as from other branches of biology (e. g., Chilton, 1909, p. 806), that the range of living forms cannot be accounted for on the supposition that the present land-masses have been permanent, i. e., the marine Algae in their present distribution tend to support such a belief in changing continental areas rather than negative it.

So far, we have been considering chiefly the east and west dis-tribution of our species. There remains to be considered the widely ranging species that occur on both sides of the Equator, and in more oceans than one. There are, of course, a number of practically cos-mopolitan species, and a larger number of genera found in all seas. Generally speaking, there seems to be—as might be expected—a rather stronger relationship with North Pacific scas, than with the North Atlantic, but this relationship is one of genera rather than species. There is a still more distinct relationship with the Indian Ocean and South Atlantic, particularly in warm-temperature and sub-tropical species. However, these remote connections are easily exaggerated, and subsequent investigations may show them slighter than now appears. Species considered to be the same in widely distant regions will be shown to be dissimilar; and this is particularly true of common European and New Zealand species. As our seaweeds have originally been examined by English and French algologists, the tendency has been to collate them with British and European species rather than with those of the North Pacific. Modern investigation, indeed, has reduced the apparent number of British species in New Zealand, so that our connection with Europe, apare from that indi-cated by cosmopolitan or widely distributed forms, is now known to be small or even non-existent. Amongst the brown seaweeds there is clearly no connection. This early became apparent. Hooker and Harvey, writing in the Flora Antarctica, vol. II., p. 45, thus discuss the distribution of the Fucoideae:

Throughout all latitudes the two tribes Fucoideae and Cys-toseireae form that prevailing vegetation to which the name seaweed is commonly applied; and the different genera so far arrange them-selves within geographical limits, as to present with such exceptions as the Scytothalia Jacquinotii, a most harmonic assemblage. Thus in the opposite frigid zones the waters are inhabited by certain genera of Fucoideae, which are in a great measure representative of one

another. The north cool zone Fucus proper and Himanthalia are represented in analogous southern zones by D'Urvillea and Sarcophy-cus. None of these genera approach the tropics, for the Fucoideae abound toward the poles, and there attain their greatest bulk diminish-ing rapidly towards the Equator, and ceasing some degrees from the line itself. The representative of the Cystoseireae in the higher lati-tudes of the opposite hemisphere, are equally appropriate with those of the Fucoideae, for we have in the North cool zone Cystoseira and Halidrys represented in the South cool zone by Blossvillaea and Scyto-thalia whilst the immense genus Sargassum find its maximum in lower latitudes and under the Equator itself.”

This passage, when considered in the light of modern ideas of classification, will have to be very much modified. Oltmanns (1922) divides the Fucales into two families the D'Urvilleaceae and the Fucaceae. In the former are placed the two genera D'Urvillea and Sarcophycus with a South circumpolar distribution. The Tahiti habitat for the former is probably erroneous. We can therefore no longer consider these two genera as representative of the northern Fucus and Himanthalia, which belong to the Fucaceae proper. The closest southern representative of Fucus is probably the Australasian genus Xiphophora, though we are still ignorant of many of the details of its structure necessary for a full classification. The section of the family to which Himanthalia belong is not represented in southern waters. Scytothalia is now also regarded as belonging to the Fucus-Ascophyllum group and not to the Cystoseireae. Further, the genus Cystoseira has recently been discovered in New Zealand (de Toni et Forti, 1923, p. 70).

Thus, it is now impossible to see the balanced arrangement in northern and southern waters that Hooker and Harvey found. Cer-tainly the northern Cystoseira seems to be largely represented by the southern Cystophora (Blossvillaea), but otherwise there is no balance. We have also the anomalous genera Hormosira and Notheia, and the endemic Landsburgia whose position is as yet insufficiently known. On the whole, therefore, it may be stated that the members of the Fuelaes found in New Zealand, though so closely related to those of Australia, show very little relationship to those of European seas. In fact, with the exception of Cystoseira, there is not a single genus common to Great Britain and New Zealand—a disparity that it would be difficult to parallel in any other equally widespread algal family. On the other hand, most of the New Zealand genera of Cholorophyceae are also found in Great Britain, and the same will probably be found to be true for the remaining Phaeophyeeae with the exception of the Laminariaceae. This perhaps tends to show that the Fucaceae is the most-recently differentiated family of the sub-class.

The general distribution of the Laminariaceae has been well studied by Setchell (1893). One of the most noteworthy facts is the complete absence of the otherwise cosmopolitan genus Laminaria from Fuegian, Australasian, and Antarctic seas. The genus Lessonia, with its headquarters in Fuegia, has found its way as far north as the Ochotsk Sea in the Pacific. It is quite absent from the Atlantic. Macrocystis is temperate south circumpolar, but also travels up the

west coasts of America and Africa. Ecklonia, with a somewhat similar distribution to Macrocystis in the south, is apparently absent from the N. W. Pacific, where, however, it is replaced by the allied genus Eisenia. On the other hand, Ecklonia is largely represented in Japan, where Macrocystis is wanting. Altogether, we may say that our Laminarians are more closely related to those of the Northern Pacific, than of the Northern Atlantic. Apparently the only genus of the Laminariales common to New Zealand and Great Britain is Chorda, and in the Laminariaceae proper, there is no common genus. As in the Fucaceae the northern and southern genera are widely different. Again, the Fucaceae of the N. W. Pacific show rather more relationship to those of New Zealand, than do the Fucaceae of Northern Europe, particularly in the Sargassum series, for the genera Blossevillea (Cystophora) Cystoseira, and Sargassum occur in both districts, though Blossevillea (Cystophora) is represented only by one rare species in N. W. America, and Cystoseira is represented by one apparently rare species in New Zealand.

A plant of remarkably discontinuous distribution, which has been the subject of some discussion may be worth mentioning here—Codium mucronatum J. Ag. This species was originally described by Agardh in 1886, and three varieties were established, depending on the char-acter of the mucro: var. Tasmanicum from Tasmania, var. Califor-nicum from California, var. Novae-zelandiae from New Zealand. The plant was subsequently recorded from South Africa, Japan, and Cape Horn; and the var. Novae-zelandiae from East Australia. Cotton (1912, p. 117), described a plant from Clare Island as C. muc-ronatum var. Atlanticum, and stated its closest ally to be the New Zealand from; and says of it (l. e. p. 118): “This remarkable feature with regard to the distribution of C. mucronatum is not so much its link with the Australian forms (our other species C. tomentosum and C. adhaerens apparently occur in the Southern Hemisphere), but its isolated position in Europe. It is not known from the Mediterranean or from North Africa, and is apparently absent from the remainder of the North Atlantic.” It is, however, found at various points in Ireland, Scotland, and the Isle of Man. Further he continues (l. e. p. 171): “Codium mucronatum var. Atlanticum, is even more note-worthy. So closely allied to the New Zealand form of the species as to be almost inseparable from it, and yet unknown in the Northern Hemisphere except in the British Isles, its distribution is certainly remarkable. As far as is known, it does not occur in England, but has existed in Scotland for at least seventy years, and in Ireland for upwards of a hundred.” However, Setchell and Gardiner (1920, p. 171), identify C. mucronatum with C. fragile (Suring) Hariot, and state: “The mucronate tip of the utricle of this plant is a prominent specific character and is subject to extreme variation. We have studied and compared plants from a considerable number of localities ranging from Alaska to Mexico; and have come to the conclusion that species cannot be split into varieties based on that character.” (i. e., of the muero.) It is certainly true that the New Zealand specimens I have examined show considerable differences in the development of the mucro, though as yet I have not seen any of the Tasmanian type.

I hope to give the matter more detailed study later on. This, of course, still leaves the British plant with a curiously discontinuous distribution.

We have still to say a few words as to our connection with N. W. America. There are several common forms which are worthy of mention. Codium mucronatum J. Ag., has already been discussed. Another remarkable case is that of Enteromorpha acanthophora. It is known from New Zealand and Tasmania, and has recently been identified from Mexico by Setchell. It was originally described by Kuetzing from New Zealand from specimens given him by Sir J. D. Hooker. However, so little is known about the Enteromorphas of the Pacific generally, and they are so difficult to identify, that little stress need be laid on this case of discontinuous distribution. Gigartina radula again would be a sub-Antarctic species except for its occur-rence in N. W. America, and Aeodes, hitherto regarded as an endemic New Zealand genus, has recently also been discovered there; but gener-ally speaking there is little connection indeed between our seaweeds and those of the N. W. Pacific, which are related to the algal Flora of the Atlantic rather than to that of the South Pacific. There is, however, a slight and fairly definite connection represented by certain red seaweeds, which are apparently southern forms that have some-how crossed the tropical barrier and reached the coasts of North West America. Such species (in addition to Aeodes and Gigartina radula, already mentioned) are: Iridaea laminarioides, I. cordata, Rhodo-glossum latissimum, Callophyllis variegata, Schizoneura (phycodrys) quercifolia, Grateloupia pinnata. Generally speaking, however, the tropics are an almost complete barrier to the passage of red and brown seaweeds from north and south temperate regionis and vice versa; yet four of these species beong to our sub-Antarctic list. We may sum up by saying that the phytogeographic relationships of the New Zea-land algal region are chiefly with Australia and through Australia to some slight extent with the Indian and South Atlantic Oceans. There is a small sub-Antarctic element, and a still smaller North Pacific one. A curious case is Caulacanthus spinellus, which is known only from Juan Fernandez, Easter Island, and New Zealand.

For the sake of students and investigators generally reference may be made to one or two of the problems confronting the New Zealand algologist. The recent discovery by Kylin, Sauvageau, and others, of a sexual generation in the European Laminarians, makes it certain that similar discoveries will in the future be made in respect of our three large Laminarians, Macrocyslis, Lessonia, and Ecklonia. There can be no question of the closeness of the relationship of these three genera. They represent the maximum of complexity of vege-tative structure in any group of Algae. The occurrence in all of them of conducting strands, sieve-tubes (trumpet-hyphae), etc., places them at the head of the brown algae; but no gametophyte is known in these three genera. Doubtless it is minute and exceedingly difficult to obtain. Probably the sexual generation is only to be obtained by cultivation in an aquarium, for the sporophytes are all large seaweeds, whose cultivation in the open would be difficult. Perhaps the only hope of dealing successfully with them would be in

a properly equipped marine biological station, and this we do not at present possess in New Zealand. It is hoped that some institution of the kind may soon be developed here; and I believe Professor Kirk contemplates the establishment of such a station at Island Bay. There would be no better place in the world for the study of the alternation of generation in these huge Laminarians, for Lessonia, Macrocystis, and Ecklonia all abound there. In the genus Lessonia (though not in the New Zealand species), huge tree-like form occur; Macrocystis is the longest seaweed in the world, often attaining a length of much over a hundred feet, and Ecklonia, though much smaller, is a comparative giant as compared with most other algae. On the other hand, it is possible that in contrast with these huge sporophytes the gametophyte consists of a few microscopic cells bearing the sexual organs.

Though the Laminariaceae have pride of place amongst the brown seaweeds, and present the most pressing problems for investi-gation, some of the other Phaeophyceae are in urgent need of further and more detailed examination. The strange plant of south temperate seas, Splachnidium rugosum (L) Grev. is still a puzzle. By the earlier algologists it was unhesitatingly placed in the Fucaceae. How-ever, in 1892 a special order was constituted for it by Miss M. Mitchell and Miss F. Whiting, who believed that they had proved that the con-ceptacles contained zoospores and not oospheres as had previously been supposed. I was fortunate enough shortly afterwards (1893, p. 288) to see these zoospores (if zoospores they be and not isogamous gametes) in free motion outside the conceptacles. The same investi-gators also found a remarkable initial cell at the apex of each branch and at the base of each conceptacle. Later Miss M. Roe (1916, p. 400) found it preferable to retain the plant among the Fucaceae, regarding it as an intermediate between that family and the Lamin-ariaceae. In 1920, Dr. Skottsberg investigated the plant and came to quite other conclusions (1920, pp. 277–287).

In order to explain his results, some account must be given of the general somatic structure in certain groups of the Phaeophyceae. In the Ectocarpaceae we have much branched filaments, consisting of single rows of cells. When we have these interwoven into ropes we get the cable type. This occurs in New Zealand in Chordaria and in Mesogloia (if the last named is really found here). Then in the Fucaceae we have the parenchymatous type, and further in all Laminarians trumpet-hyphae are present. Now Miss Mitchell and Miss Whiting saw in Splachnidium an alga of the parenchymatous type with zoospores in conceptacles. Skottsberg re-examining the plant in 1920, finds the thallus filamentous and pseudoparenchymatous; and feels justified in regarding it as of a Chordariaceous type. In Chordaria the ultimate ramuli bear the reproductive organs on their surface, and therefore Splachnidium is quite unique in the group—if it should be retained here—in possessing conceptacles.

Now a word or two about the so-called initial cells. These have not been seen to divide, nor do they in any way appear to be connected with the rest of the plant. Dr. Skottsberg regards them—and appar-