Search

Tmesipteris horomaka sp. nov. is a rare octoploid, endemic to Banks Peninsula, New Zealand. It is morphologically intermediate between the tetraploids Tmesipteris elongata and Tmesipteris tannensis...

The holotype of Tmesipteris fowerakeri H.N. Barber is identified and lectotypes are chosen for two basionyms relevant to New Zealand representatives of the fern family Psilotaceae—Bernhardia novae-hollandiae Mull.Berol. and Psilotum heterocarpum Colenso. This article is a contribution towards clarifying the taxonomic and nomenclatural status of New Zealand plants for the plant names database (Ngā Tipu Aotearoa) and the electronic Flora of New Zealand.

This article describes the results of a light and electron microscopic study of the fungal endophytes and vascular anatomy in the rhizomes and gametophytes of Tmesipteris and Psilotum. The parenchymatous cortical cells of the rhizomes and subterranean gametophytes of Tmesipteris and Psilotum contain intracellular aseptate glomeromycotean fungi resembling the “Paris-type” of arbuscular mycorrhizas found in seed plants. The fungi differentiate into multinucleate vesicles and hyphal coils, both containing bacteria-like structures and accumulating lipid masses and crystals as they age. After several cycles of infection in the same cell, degenerate hyphae form amorphous masses encased by host wall material. Nearly identical host–fungus cytology between the autotrophic sporophytes and the heterotrophic gametophytes suggests that these psilophyte associations are exploitative of the fungus in both generations. Following the description of tracheids nearly 60 years ago in the gametophytes of Psilotum, vascular elements are described for the first time in the haploid generation of Tmesipteris. Close similarities between the water- and food-conducting elements in both generations, viz. vessel elements with scalariform perforation plates and sieve cells with refractive spherules and lacking callose at all stages in their develoment, add support to the homologous theory of the alternations of generations. Mitochondrial aggregations, cross-linked by small electron-opaque rods, are common in the stelar cells of both generations and appear to be a unique feature of the psilophyte clade.

This light and electron microscope study revealed that leptosporangiate ferns have highly distinctive gametophyte‐sporophyte junctions characterized by sporophytic haustoria, the absence of intraplacental spaces and degenerating cells, and the early appearance of wall ingrowths in both generations. Other notable cytological features are highly pleomorphic plastids and mitochondrial aggregates in the gametophytic placental cells. Close similarities with the gametophyte‐sporophyte junctions in Tmesipteris and major differences from those of homosporous lycophytes are in line with the placement of psilophytes and ferns in the same clade and distance both from lycophytes. A smooth interface between the two generations in Azolla suggests a clear-cut discontinuity between homosporous and heterosporous ferns, although this is the only heterosporous fern investigated to date. Similarities between the gametophyte‐sporophyte junctions of leptosporangiate ferns and hornworts, when balanced against differences between them, are considered more likely the result of parallel evolution rather than homology. a 2003 Annals of Botany Company

We examined collections of pteridophytes from Moorea and others of the Society Islands, as well as literature relevant to the pteridophytes of Polynesia. This resulted in a list of species known to occur on Moorea, along with a list of species reported for Moorea but lacking voucher specimens, and a list of species perhaps to be found on Moorea based on collections from nearby Tahiti and adjacent islands in the archipelago, at suitable elevations. We include habitat, locality, and appropriate taxonomic commentary for each known species. A new species in the family Psilotaceae, Tmesipteris gracilis Chinnock, is described from the Society and Marquesas Islands. We also include a discussion of pteridophyte collection history on Moorea and biogeographic notes for species on the island.

We report the chloroplast genomes of a tree fern (Dicksonia squarrosa) and a “fern ally” (Tmesipteris elongata), and show that the phylogeny of early land plants is basically as expected, and the estimates of divergence time are largely unaffected after removing the fastest evolving sites. The tree fern shows the major reduction in the rate of evolution, and there has been a major slowdown in the rate of mutation in both families of tree ferns. We suggest that this is related to a generation time effect; if there is a long time period between generations, then this is probably incompatible with a high mutation rate because otherwise nearly every propagule would probably have several lethal mutations. This effect will be especially strong in organisms that have large numbers of cell divisions between generations. This shows the necessity of going beyond phylogeny and integrating its study with other properties of organisms.

Perforation plates are reported in aerial and subaerial axes of Psilotum nudum and in aerial axes of Tmesipteris obliqua. In Psilotum, both perforations lacking pit membranes and perforations with pit membrane remnants were observed. Perforation plates in Psilotum may consist wholly of one type or the other. In Tmespteris, perforations have threadlike pit membranes or consist of porose pit membranes. Wide perforations alternating with narrow pits, a conformation observed in various ferns, were observed in Psilotum (subaerial axes). In Psilotum, perforations are more common in metaxylem than in protoxylem; perforations in protoxylem consist of primary wall areas containing small circular porosities or relatively large circular to oval perforations. There are no modifications in the secondary wall framework of protoxylem or metaxylem in Psilotum or Tmesipteris that would permit one to distinguish presence of perforations or perforation plates with light microscopy, and scanning electron microscopy (SEM) is required for demonstration of porose walls or perforations. The tracheary elements of the Psilotaceae studied have no features not also observed in other ferns with SEM.

Spores of several species of Psilotum and Tmesipteris were sown on a nutrient medium containing minerals and 0.2% glucose. The nutrient medium contained ammonium as the nitrogen source and lacked nitrate. From 52 to 98% of the spores of P. nudum, P. complanatum, T. lanceolata, and T. sigmatifolia germinated after 6 months in the dark. Haploid spores of P. nudum began germinating in less than 2 months; however spores of the two Tmesipteris species reached 50% germination sooner than those of the Psilotum species. Spores of T. tannensis never germinated and those of T. elongata and P. × intermedium rarely germinated (

6.2. Tmesipteris Bernh., J. Bot. (Schrader) 1800(2): 131 (1801). T.: Tmesipteris tannensis (Spreng.) Bernh. (Lycopodium tannense Spreng.), Published as part of Christenhusz, Maarten J. M., Zhang, Xian-Chun & Schneider, Harald, 2011, A linear sequence of extant families and genera of lycophytes and ferns, pp. 7-54 in Phytotaxa 19 on page 41, DOI: 10.11646/phytotaxa.19.1.2, http://zenodo.org/record/4893995

Two principal types of shoot apex are recognized, (1) the classical pyramidal apical cell of pteridophytes which does not divide periclinally, and so never contributes directly to the inner tissues of the axis, and (2) that of seed plants, in which there are one or more superposed initials of which the innermost contributes directly to inner tissues by periclinal divisions. The occurrence of these types in the major groups of living land plants is reviewed. The first type is present in ferns,Equisetum, Psilotum, Tmesipteris, Selaginella, and bryophytes (when apical growth occurs in them). The second type occurs in all angiosperms and gymnosperms, and also inLycopodium, Phylloglossum, Isoetes, andStylites. No exceptions nor intermediate conditions occur. Both types of apex are known in Carboniferous fossils, the first type in sphenopsids and the second in pteridosperms and cordaitean gymnosperms. These apices define two evolutionary lines of land plants which may have originated separately as sporophytes from ancestors with no diploid generation. Correlation of these types with the following features is considered: sieve-element inclusions; initiation of leaves and sporangia; the primary vascular system; embryogeny, the vascular cambium; initiation of lateral roots and the type of branching. Seed plants and pteropsids differ in respect of all these features, giving support to the hypothesis that the sporophytes of these two groups arose independently. The position of sphenopsids and lycopsids is also considered.

Chloroplast 16S rDNA sequences from the charophyte algae Spirogyra and Coleo- chaete; the bryophytes Marchantia and Sphagnum; the fern allies Equisetum, Isoetes, Selaginella, Lycopodium, Tmesipteris and Psilotum; the eusporangiate ferns Angiopteris, Botrychium, and Ophioglossum; the leptosporangiate ferns Lydogium and Doodia; the gymnosperms Ephedra and Juniperus; and six angiosperms (two monocots and four dicots) were analyzed with the parsimony algorithm PAUP to infer phylogenetic relationships. The charophytes were the designated out- group and the bryophytes were at the base of the land plant clade. Relationships of most of the fern allies and Angiopteris were unresolved but a clade containing Psilotum + Tmesipteris (Psi- lotales) was sister to a clade composed of Botrychium and Ophioglossum (Ophioglossales). A leptosporangiate fern clade was strongly supported, and this clade grouped weakly with Angiop- teris. The Selaginella 16S rDNA sequence is apparently changing at a faster rate than the other sequences as evidenced by the extremely long branch associated with Selaginella. The effect of this long branch on the overall tree topology was tested by removing Selaginella from a subsequent analysis. Trees produced without Selaginella weakly supported the placement of Lycopodium at the base of the vascular plants clade but Isoetes grouped weakly with the seed plants. Removal of Selaginella also resulted in higher bootstrap values and decay indices for the majority of the clades. The determination of phylogenetic relationships of the "fern allies," the Psi- lophyta, Lycophyta, and Sphenophyta, has proven to be elusive. Taxonomi- cally, the fern allies are generally considered to be positioned between the bryophytes and ferns but formal analyses of morphological and molecular data sets only partially resolve their relationships. In a cladistic analysis of mor- phological characters, Bremer et al. (1987) placed Psilotum at the base of vas- cular plants, the lycopsids branched off next, then Equisetum, and then the ferns. However, the portion of the clade containing the fern allies was con- structed with only four morphological characters and it is not reasonable to expect that a robust phylogeny containing highly divergent taxa can be pro- duced from such a small data set. Conversely, fossil evidence (Knoll and Roth- well, 1981; Stewart, 1983; Gensel and Andrews, 1984), characters relating to male gametogenesis (Garbary et. al., 1993), and chloroplast genome structures (Raubeson and Jansen, 1992) support the hypothesis that the lycophytes are at the base of the vascular land plant clade. Mishler et al. (1994) analyzed green plant molecular and morphological data sets separately and together, using parsimony algorithms. Some vascular land plants were included, with the fern allies represented by Equisetum, Lycopo- dium, Lycopodiella, and Selaginella. The analysis of "general" morphological characters placed the lycophytes sister to a clade containing Equisetum, ferns

The flavonoids of the primitive leptosporangiate ferns Stromatopteris moniliformis, Schizaea bifida, Gleichenia cunninghamii, Cardiomanes reniforme, and Hymenophyllum demissum have been identified as 3-0-glycosides of the flavonols kaempferol and quercetin. None of the examined ferns produced flavonoids which are also common to the Psilotoceae. The Psilotaceae have previously been shown to produce 0-glycosides of amentoflavone (biflavone) and apigenin (flavone) and traces of C-glycosylflavones (Wallace and Markham, 1978). These data imply that the primitive leptosporangiate ferns are not closely related to the Psilotaceae. The antiquity of rhamnose as a glycosyl moiety is suggested by its occurrence in both the primitive leptosporangiate ferns and in the Psilotaceae. SEVERAL interpretations of fern phylogeny have been published (Pichi-Sermolli, 1973). All have considered the Schizacaceae, Gleicheneaceae, Stromatopteridaceae, and Hymenophyllaceae as representing primitive leptosporangiate ferns. Most previous authors considered the Psilotaceae to represent a distinct and only remotely related taxon. However, Bierhorst (1968, 1977) challenged the position of the Psilotaceae as being a well defined entity at the divisional level. He reclassified the Psilotaceae, containing both Psilotum and Tmesipteris, into the Pteridophyta where it was closely allied with the monotypic Stromatopteridaceae. He supported this by detailed morphological studies involving ontogeny of both sporophytes and gametophytes. Most convincing were the anatomical comparisons of young sporophytes of Psilotum, Tmesipteris and Stromatopteris as well as his observation of an "annulus" on the psilotaceous synangium. However, as evidenced at a recent symposium, many morphologists are not in agreement (White, 1977; Gensel, 1977; Kaplan, 1977; Wagner, 1977). Subsequently Cooper-Driver (1977) presented phytochemical evidence that the Psilotaceae should not be allied with Stromatopteris or any other primitive (or advanced) fern. She contended that I Received for publication 28 March 1978; revision accepted 5 July 1978. The authors are grateful to Dr. Phillip Morat, Noumea, New Caledonia, and to Dr. Patrick Brownsey, The National Museum, Wellington, New Zealand for supplying Stromatopteris moniliformis. J.W.W. acknowledges the support of a New Zealand National Research Advisory Council Senior Research Fellowship and a Faculty Research Grant from Western Carolina University. the presence of amentoflavone as the only flavonoid in both Psilotum and Tmesipteris with the corresponding absence of flavonols was justification to keep the Psilotaceae as a separate entity at the divisional level. Recently Wallace and Markham (1978) documented the total flavonoid profile of species of Psilotum and Tmesipteris. They identified amentoflavone and its 0-glycosides as the major components of both genera and reported the presence of flavone-O-glycosides as minor constituents with C-glycosylflavones occurring as trace components. Although amentoflavone and its 0-glycosides have not been documented in ferns, it is known that some ferns have the ability to synthesize flavone 0-glycosides (Swain and Cooper-Driver, 1973) and C-glycosylflavones (Ueno et al., 1963; Soeder and Babb, 1972). Interestingly, a full analysis of the flavonoid glycosides of Stromatopteris and other primitive leptosporangiate ferns has not as yet been published. The present paper is a detailed account of the flavonoids of Stromatopteris moniliformis, Schizaea bifida, Gleichenia cunninghamii, Cardiomanes reniforme, and Hymenophyllum demissum. Phylogenetic implications of these compounds pertaining to the relationship of the primitive leptosporangiate ferns to themselves and to the Psilotaceae will be discussed. MATERIALS AND METHODS-Plant materialsStromatopteris moniliformis Mett. (14.0 g d.w.) was collected in March 1977, at Noumea, New Caledonia; Schizaea bifida Willd. (1.8 g d.w.) was collected on 12 April 1977, Auckland, New

Tmesipteris Bernh. is revised in New Zealand including the Kermadec, Chatham, and Auckland Island Groups and four species are recognised, namely T. tannensis (Spreng.) Bernh., T. elongata Dang., T. lanceolata Dang., and T. sigmatifolia sp. nov. In addition, Tmesipteris elongata is further subdivided and a new subspecies, ssp. robusta, is proposed. Detailed descriptions of the species are supplied and their distributions and ecology discussed. Notes on other species are given and features which are considered useful in the grouping of species are introduced, although infrageneric subdivisions are not proposed.

Chromosome counts of n = 104 are reported for New Zealand populations of Pstlotum nudum. Tmesipteris elongata. T. lanceolata. T. Slgmatifolia and T. tannensis. These results confirm that Psilotum and Tmesipteris share the same base number, x = 52, and show that the New Zealand species of Tmesipteris are all tetraploid compared with Australian species which include both tetraploids and octoploids. A difference in ploidy supports the morphological dlstinction between the New Zealand T. tannensis and the closely related Australian T. billardierei. The recent reinstatement of Psilotum heterocarpum Colenso as a distinct species is shown to be based on a misconception. Cytological observations relevant to the hypothesis that the Psilotaceae are related to primitive members of the Filicales are discussed. No substantial cytological evidence can be found to support this concept which is also contrary to phytochemical and other evidence. The origin and relationships of the Psilotales remain open questions...

Documentation of amentoflavone O -glucosides as the predominant flavonoid glycosides in both genera of the Psilotaceae clearly distinguishes this family from all other families of vascular plants. Psilotum and Tmesipteris also possess apigenin C - and O -glycosides as common flavonoid types. Apigenin 7- O -rhamnoglucoside occurs in both genera and the previously undocumented apigenin 7- O -rhamnoglucoside-4′- O -glucoside, although identified only in Tmesipteris , may also be present in Psilotum . The existence of flavone C -glycosides in both genera may provide a phytochemical relationship between the Psilotaceae and some ferns. The phylogenetic significance of these results is discussed.

Summary When Bernhardi established Tmesipteris in 80oi he based it on Lycopodium tannense Spreng., described the previous year from a collection supposedly made on Tanna Island (New Hebrides) by the Forsters. The holotype of L. tannense, the type species of Tmesipteris, has not been located and it is presumed destroyed. Tmesipteris is now generally considered polyspecific and since the protologue account of L. tannense is only sufficient to place the species in Tmesipteris further identification, to distinguish it from other species now recognized, is necessary. As no isotype of L. tannense has been located the figure in Bernhardi's paper, which is considered to depict the type collection, is selected as neotype. The type collection of L. tannense was stated by Sprengel to have been collected on Tanna Island. However, the morphological, biogeographical and historical evidence suggests that the collection was made in New Zealand, probably at Dusky Sound. Tmesipteris tannensis (Spreng.) Bernh. is now considered to be restricted to New Zealand, and the reference by Sprengel to Tanna Island is considered an error.

Each family in this paraphyletic group contains a characteristic array, occasionally with unique compounds, of flavonoid compounds. The Psilotaceae have amentoflavone and its glucosides as well as a few flavone O- and C-glycosides. The Lycopodiaceae contain an array of flavones which may be of systematic value. The Selaginellaceae contain amentoflavone and hinokiflavone and their methyl ethers, as well as some C-glycosylflavones. The Isoetacae are poorly known but appear to contain mainly flavones. Lastly, the Equisetacae contain mainly flavonols but a diversity of compounds have been reported from one species. Proanthocyanidins appear for the first time in the living land plants in Equisetum. The only other novel compounds are the biflavone glucosides in Psilotum and Tmesipteris and a 2′-hydroxylated flavone in Isoetes. There are apparently no derived compounds to link together any of the families, with the possible exception of some uncommon flavones in the Lycopodiaceae and the Isoetaceae. Taxa worthy of further investigation include Lycopodium, Isoetes and Selaginella.

The covalently bonded components of the stem cutin of Psilotum include 16-hydroxyhexadecanoic acid and substantial amounts of hexadecane-1,8,16-triol. While of generally similar composition, leaf cutin of Tmesipteris contains a mixture of hexadecanetriol isomers. The findings suggest that psilotophyte cutins evolved in a different manner from those of other land plants.

A B S T R A C T The external spore morphology of Phylloglossum drummondii was studied with the scanning electron microscope and shown to share a number of features with spores in the subgenus Urostachys in Lycopodium. Within Urostachys the affinities of P. drummondii were less evident: the foramino-fossulate distal face, pyramidal proximal surface, and subcircular outline of its spores are similar to spores in the section Phlegmaria while the weakly foveolate proximal surface and raised laesurae are characteristic of spores in the section Selago. IT HAS been generally accepted by a majority of workers that the affinities of Phylloglossum drummondii Kunze are with the genus Lycopodium L. (for example see Bierhorst, 1971; Bower, 1885, 1935; Eames, 1936; Foster and Gifford, 1959; Hackney, 1950; Pritzel, 1900; Rothmaler, 1944; Smith, 1938; Sporne, 1966; Walton and Alston, 1938). The two genera form the extant element of the Lycopodiaceae Rchb. which is typified by eligulate microphylls and homospory, with a single sporangium adaxially attached to or associated with each sporophyll (Foster and Gifford, 1959; Pritzel, 1900; Walton and Alston, 1938). Two differing points of view are those of Worsdell (1910) and Wernham (1910). Worsdell considered the rhizophore of Selaginella Beauv. homologous with the perennating tuber of Phylloglossum, while Wernham saw similarities between the vascularization of Phylloglossum and Tmesipteris. Neither view has gained general acceptance. Although its validity has been questioned (Pritzel, 1900; Rothmaler, 1944), Phylloglossum has been maintained as a distinct, monotypic genus by them and others subsequent to Kunze. However, as Eames pointed out, the criteria of the unbranched habit and leaf arrangement used by Kunze in 1843 to characterize the genus are at best dubious. Rather, the siphonostelic vascular system with mesarch xylem combined with the "extreme geophytic specialization in habit" serve best to distinguish Phylloglossum from Lycopodium (Eames, 1936). A major difficulty has been that of discerning the affinities of Phylloglossum within the heterogeneous taxon Lycopodium. Much of the problem appears to be due to difficulties in interpreting the extremely reduced structure and anomolous growth habit of P. drummondii. Bower (1885, 1935) incorrectly interpreted

REEVE, R. M. 1942. Structure and growth of the vegetative shoot apex of Garrya elliptica Dougl. Amer. Jour. Bot. 29:679-711. . 1943. Comparative ontogeny of the inflorescence and the axillary vegetative shoot in Garrya elliptica. Amer. Jour. Bot. 30:608-619. . 1948. The "Tunica-Corpus" concept and the development of shoot apices in certain dicotyledons. Amer. Jour. Bot. 35:65-75. R6SLER, P. 1928. Histologische Studien am Vegetationspunkt von Triticum vulgare. Planta 5:28-69. SACHS, J. VOI. 1882. Textbook of Botany. 2nd ed. Trans. by S. H. Vines. Oxford. ScHMIDT, A. 1924. Histologisehe Studien an phanerogamen Vegetationspunkten. Bot. Arch. 8: 345-404. STERLING, C. 1945. Growth and vascular development in the shoot apex of Sequoia sempervirens (Lamb.) Endl. II. Vascular development in relation to phyllotaxis. Amer. Jour. Bot. 32: 380-386. . 1946. Growth and vascular development in the shoot apex of Sequoia sempervirens (Lamb.) Endl. III. Cytological aspects of vascularization. Amer. Jour. Bot. 33:35-45. THODAY, D. 1922. On the organization of growth and differentiation in the stem of the sunflower. Ann. Bot. 36:458-510. * 1933. Some physiological aspects of differentiation. New Phytol. 32:274-287. VAROSSIEAU, W. W. 1940. On the development of the stem and the formation of leaves in Coffea-species. Ann. Jard. Bot. Buitenzorg 50:115-198. WARDLAW, C. W. 1924. Size in relation to internal morphology. No. 1. Distribution of the xylem in the vascular system of Psilotum, Tmesipteris and Lycopodium. Trans. Roy. Soc. Edinburgh 53: 503532. . 1925. Size in relation to morphology. No. 2. The vascular system of Selaginella. Trans. Roy. Soc. Edinburgh 54: 281-308. . 1947. Experimental and analytical studies of Pteridophytes. X. The size-structure correlations in the filicinean vascular system. Ann. Bot. N.S. 11: 203-217. YARBBROUGH, J. A. 1934. History of leaf development in Bryophyllum calycinum. Amer. Jour. Bot. 21: 467-484. ZIMMERMANN, W. A. 1928. Histologische Studien am Vegetationspunkt von Hypericum uralum. Jahrh. Wiss. Bot. 68:289-344.

Although the vascular system of the leaf in the Cycadaceae has for some time been recognized as a region where primitive structures, notably centripetal wood, have persisted longer than anywhere else in the plant, the general principle of the conservatism of the foliar organ has only recently been more widely extended and found to hold good for all the main groups of vascular plants: Lycopsida, Filicales, Gymnosperms, and Angiosperms. The position of the protoxylem in relation to the later formed elements of the wood is very constant in the large orders, but its importance in determining phylogenetic relationships seems often to have been overlooked. A brief summary of the structure of the leaf trace and foliar conducting system of all families of vascular plants, with special reference to the position of the protoxylem, and a more particular account of conditions in the living Cycadaceae, the only existing gymnosperms with centripetal wood, together with a discussion of the bearing of the results on our views as to the evolutionary history of the higher plants, is the purpose of the present paper. That the base of the leaf trace, since it is least subject to changes in external conditions, should retain ancestral structures longer than does any other portion of the foliar system seems logical, and this has been proven to be true in several groups of vascular plants. It is therefore in this region that we should look for the primitive structure of the foliar bundle. In all the Lycopsida where the anatomy of the leaf is known (Equisetum, Lycopodium, Selaginella, Phylloglossum, Tmesipteris, Psilotum, Isoetes, Lepidodendron, and Sigillaria), the vascular strand at the base, and in the majority of instances throughout the lamina, of the leaf, is a single concentric strand, roughly circu

Considerable interest attaches to the unravelling of the complete life-history of the Psilotaceæ, since they are apparently the nearest living representatives of the Sphenophyllales.The recent description of the gametophyte of Tmesipteris by Prof. Lawson, whose figures 1 had the privilege of seeing before they went to press, and the discovery of the gametophyte of Psilotum, which is the subject of the present paper, form a first contribution to the solution of this problem. It is of some interest to note that there remains now no Pteridophyte the gametophyte of which has not been described.In the early part of this research I was fortunate in finding a locality near at hand in which Psilotum flourished, and therefore I had at my disposal abundance of spores and later numerous prothalli.

In his Presidential Address to the Royal Society of Edinburgh in October 1920, Professor Bower (2) called attention to the question of Size in relation to stelar morphology, and advanced evidence (with particular reference to the Filicales) to show how the principle of similar structures has affected the internal morphology of the vascular system of plants. In dealing with the application of this principle he remarks that “the stems and roots of most plants are approximately cylindrical. The same is the case as a rule for their conducting tracts also. The cylinder is one of those solid forms in which the proportion of external surface to bulk is exceptionally low. Any deviation from the cylindrical form, either by external projections or by involutions, necessarily leads to an increase in the proportion of surface to bulk. The surface varies only as the square of the linear dimensions, but the bulk as the cube. It follows, therefore, that in carrying out any of those physiological functions of a living organism which depend on surface, as do all those of the acquisition and inter change of material, the actual size of the part which exercises that function is a matter of the greatest moment. The larger the plant is, the more dependent will it then be upon its form and detailed structure, not only for its stability, but also for the performance of its functions of absorption and transit of liquids and gases. This will apply not only to the external surface, but also to those internal surfaces which limit one tissue tract from another.” In illustration of this, examples were taken from the stems, tubers, and petioles of the Filicales, and the size-factor was also found applicable to the large prop-roots of palms, such as Areca and Verschaffeltia.

The peculiar fertile structures of Tmesipteris and Psilotum have given rise to much discussion as to their morphological nature. The foliage leaves in Tmesipteris, as is well known, are simple, whereas the fertile structures are forked, that is, each has a stalk bearing leaf-like lobes, a bilocular sporangium or synangium being attached to the stalk just below the lobes. Three views have been taken as to the nature of these fertile structures— 1. The whole structure has been interpreted as a branch of the stem, being equivalent to a reduced Lycopodium spike. The lobes, according to this view, will be leaves. (Juraniyi, Strasburger, Sachs, Goebel.)

In some slides from young synangia of Tmesipteris , prepared from curiosity regarding this unique plant, it was noticed that during mitosis there were present, in addition to the chromosomes, several rounded bodies similar to the nucleoli of resting cells. Moreover, nucleoli were present in nuclei where the spireme was fully developed prior to mitosis. Preliminary examination thus suggested Tmesipteris as a most favourable plant for a study of the nucleolus as regards its relationship to the chromosomes and its behaviour during mitosis. Historical . The study of plant-nucleoli has been largely incidental to the investigation of chromosome behaviour, more especially during the meiotic divisions. The general conclusions from the study of plant-nucleoli may be summarised thus:— The number of nucleoli per cell may vary from one up to eight or ten, some of which may arise by the division of pre-existing ones. (1) In some cases they have been regarded as stores of material which contribute to the building of the chromosomes in prophase (2, 3), while in other cases this, explanation has been rejected. Another view, supported by Strasburger and Chamberlain (4, 5) amongst others, is that spindle-formation is a function of the nucleoli. Others again hold that extruded portions of the nucleolus are converted into organs such as the blepharoplast of Hepaticœ . Extrusion of chromidia into the cytoplasm by nucleoli has been described by various cytologists (6, 7), and Digby (1) finds that the nucleolus of Galtonia is closely related to one pair of chromosomes, and at metaphase is “pushed” away from the equatorial plate on the end of a chromosome. Lack of genetic continuity between nucleoli has been advocated by many authors, especially by Gates (8, 9, 10) and by Mottier (11).

The Psilotaceæ constitute one of the most interesting of existing Pteridophyte groups. This interest is mainly due to their phylogenetic isolation. They are perhaps the most isolated of existing types—showing no close affinity to other Pteridophytes. They are very highly specialised in their anatomy, habit, and habitat, and limited to a comparatively narrow geographical distribution. They include two probably monotypic genera—Tmesipteris and Psilotum—which are closely related. Although both genera are limited to the tropics and sub-tropics, Psilotum has a much wider distribution than Tmesipteris. The latter is confined to the South Sea Islands, Australia, New Zealand, and parts of Polynesia. Up till recent years they have been classed with the Lycopodiales, but this has been more a matter of convenience than an indication of close relationship, and has only served as a temporary classification until more knowledge has been obtained regarding them. The peculiar nature of these plants, as shown in their structure and in their habitat, suggests that they are highly specialised remnants of a much larger group which has practically become extinct. Recent inquiry into the structure of the sporophyte of both plants has made this much more than a suggestion. As a result of the careful investigations of Scott (1908), Boodle (1904), Bower (1894-1908), Ford (1904), and others, there has been a marked tendency to remove the Psilotaceæ from their temporary position among the Lycopodiales and associate them with the extinct Sphenophyllales. Scott (p. 631) states that it seems best to regard the Psilotaceæ as forming a class of their own, the Psilotales, under the main division Sphenopsida—their closest affinity under this class being the Sphenophyllales. This probable affinity of the Psilotaceæ to such an ancient and long extinct race as the Sphenophylls, and the great uncertainty of their relationship to other existing types, have awakened a great interest in these two specialised and isolated genera. When, in addition to this, we consider that the Psilotaceæ is the only group of Pteridophytes in which the gametophyte generation and embryo are not yet definitely known, our interest in these plants becomes twofold, and any new information throwing additional light on their life-histories will be welcomed.

Tmesipteris Vieillardi , Dangeard, is one of those rare plants from the Pacific region, of which our knowledge has remained comparatively incomplete, owing chiefly to the difficulty of obtaining adequate material. As is well known, the genus Tmesipteris occurs under several forms, which Dangeard described as so many distinct species, but which are generally grouped under the name Tm. tannensis , Bernh. The usual habit of plants belonging to this genus is that of semi-erect or pendulous epiphytes, although specimens are frequently seen growing on the ground. But it is not commonly known that one of the forms is normally erect and terrestrial, only rarely growing as an epiphyte. In its external morphology and anatomy also Tm. Vieillardi shows some noteworthy points of difference from the remaining forms ( see Plate 5, figs. 1, 5, 8 and 10), and it is not unlikely that it represents the parent type from which the other forms may have arisen. The chief object of the present paper is to describe these distinctive features of Tm. Vieillardi and to discuss their bearing upon the affinities of this plant, both within the family Psilotacese and with other Pteridophytes. Advantage has been taken of this opportunity to publish photographs of the different “forms” of the genus, so as clearly to illustrate, for what they are worth, the individual differences in habit and external morphology.

Since the preliminary announcement of the discovery of the prothalli of Tmesipteris and Psilotum communicated to the Royal Society of Edinburgh in June 1916, a considerable amount of additional material has been found, and it seems therefore desirable to place on record a more detailed account of these interesting structures. The preliminary description, as published, was based upon but a few specimens, and these were not sufficient to reveal all of the vegetative and reproductive features, nor enough to justify any broad general conclusions as to the phylogenetic position of the Psilotaceæ. The discoyery, however, has filled in an important and interesting gap in our knowledge; for the Psilotaceæ are the very last of the known Pteridophytes to surrender their gametophytes to the light of science.

1. True foliar gaps occur immediately above their corresponding leaf-traces and are not lateral to the leaf-traces. 2. Foliar gaps are absent in Phylloglossum, although a superficial examination of the anatomy of this genus might lead to the conclusion that the perforations in the tubular stele, which are sometimes lateral to the outgoing leaf-traces, are to be regarded as true foliar lacunae. 3. Foliar gaps are likewise absent in Tmesipteris, and recent statements as to their presence are based on misinterpretation or misconception. Perforations in the stele are here also found sometimes lateral to one or more leaf-traces, but these cannot be regarded as true foliar gaps. 4. Foliar gaps are absent in the Lepidodendreae and the Sigillariae, but in the more modern species of the latter perforations of the tubular central cylinder are sometimes found, which have the same relations and are susceptible of the same explanation as are the similar perforations in Phylloglossum and Tmesipteris. 5. Foliar gaps are...

Spores of Tmesipteris elongata Dang, germinated in axenic culture on a nutrient medium containing mineral elements and 0.2% glucose after 8 months in the dark. No spores germinated in cultures kept in the light. Initiating germination, the monolete laesura (scar) split in the middle. As the laesura ruptured to its ends, the cell bulged out. The first division was perpendicular to the polar axis of the spore and formed distal and proximal cells. At one end of the proximal cell, brown materials accumulated in the wall. Transfer of the two-celled gametophyte to new media did not support further growth.

The distribution of flavonoid compounds in Psilotum, Tmesipteris, and primitive filicalean ferns shows that it is unlikely the two groups are closely related. The Psilotaceae synthesize only amentoflavone and related biflavonyls which are totally absent from ferns. The primitive ferns contain the full complement of flavonols and proanthocyanidins, as found in all other filicalean families.