The islands of Fiji, Samoa and Tonga constitute a region related phytogeographically to one another. About one-third of the elements of the Tongan flora are not distributed beyond the region cited above. There are the species found both in Tonga and in the Continental Asia : about three-fifths of the pteridophytes and a quarter of the flowering plants known in the former islands. Most of the species found in Tonga have their relatives among the Asian-Malaysian elements. A little more than 10 % of our species are endemic to the Tongan Islands. Most of them seem to have differentiated on the island of 'Eua, the oldest of the islands. There are smaller numbers of species which are common between Tonga and New Zealand or Australia.

View full abstract

The families and subfamilies of the Leptosporangiate ferns (Hydropteridales excepted) are arranged on a graph (Fig. 1) according to their interrelation. For the disposition three main characters have been taken up first, namely the spore-form (bilateral or tetrahedral), the patterns of spore-germination (centrifugal, tangential or centripetal) and the types of the vascular construction (protostele, solenostele or dictyostele). Secondly, all the characters concerning both on the vegetative and reproductive organs have been regarded, some of which are picked up in Fig. 2. A list of taxa on the opposite page of Fig. 1 is a system of fern classification newly proposed by the author.

View full abstract

ヒメシダ群のいろいろな分類学的形質のうち,特に venation をとり上げ,その分類学的意義を追跡してみた.その結果,この形質は,系統群を大別する標徴になるほど絶対的なものではなく,ヒメシダ群という一つのまとまりのうちでもいろんな型が導かれているらしいことがはっきりしてきた.更に外見上類似しているようにみえる venation の各型相互間にもさまざまな関係が類推され,この形質を非常に重くみていたヒメシダ群のこれまでの分類系に対する疑問は益々深められてきた.ヒメシダ群を遊離派と網状派の差でヒメシダ属とホシダ属に弁別する分類は系統的に自然なものではなく,この群の分類の再編成は不可欠である.また,アミシダ属,アミシダモドキ属,ミドシダ属は,venation の差から類縁の乏しいものと見做されていたが,これらは相互に近縁のものであり,むしろヒメシダ群の他のものから独立したものではないかということが充分に推定される.以上のようなことから,ヒメシダ群では,venation の差は属よりもっと低い段階の差異を指標するものと考え,Stegnogramma, Thelypteris, Meniscium の三属に分類し,各属を細分するやり方が好ましいものと考える.

View full abstract

On the system of Lophoziaceae, the writer prefers to take the following view : 1) From the point of systematical view, Lophozia subgen. Massula is a taxon equivalent to such taxa as Barbilophozia, Leiocolea and Isopaches. Therefore, if Massula is remained to a subgeneric rank, the others should also be reduced to subgenera. 2) Hattoriella is hardly worthy of generic status and is to be regarded as a member of Lophozia s. lat. or of Leicolea. It shows a close affinity to Leiocolea especially in the characters of cells and oil-bodies. 3) It is reasonable to combine Sphenolobus with Anastrophyllum, for an almost complete series of intermediates is recognizable between these two genera. Eremonotus should be treated as a genus distinct from Anastrophyllum. It may be allied to Cephaloziella in the small size of plants, the form of leaves, the very small homogeneous features of oil-bodies, etc. 4) Anastrophyllum yakushimense is a peculiar species in the ecological and morphological characters. Its peculiarity possibly deserves to be given a new genus. The writer refrains here from establishing a new genus, because it is necessary to study such tropical species as A. schizopleurum, A. cucullifolium and A. integerrimum which all resemble A. yakushimense.

View full abstract

1. Calostoma japonicum H_<ENN>, which is very commonly found in Japan, is very near C. ravenelii (B_<ERK>.) M_<ASSEE>. They are, however, separated as follows. C. japonicum : Rooting stem shorter than the head or hardly exceeding the double length of it, narrower than the latter; exoperidium gelatinous in wet stage; endoperidium clay-coloured; spores oblong with somewhat larger and irregular pits. C. ravenelii: Rooting stem 3-5 times longer and commonly broader than head; exoperidium not gelatinous ; endoperidium clay to brown coloured ; spores oblong, mingled with globose one, with minute pits. 2. L_<LOYD>, Y_<ASUDA> and others have enumerated the same Japanese fungus in the name of C. ravenelii. The true C. ravenelii has not yet been found in Japan. 3. Dr. H. H_<ARA> collected two specimens of Calostoma in Bakkim (ca. 3000 m) of Sikkim-Himalaya during the Japanese Botanical Expedition (1960). This is surely C. junghuhnii (S_<CHLECHT>. et M_<UELL>.) M_<ASSEE> as already been reported by B_<ERKELAY> from Himalaya basing on the H_<OOKER>'s collection. 4. Calostoma formosanum (S_<AWADA>), which was found in Formosa, seems to be a form of C. junghuhnii with very short rooting stem.

View full abstract

1. Japan-Hondo (Honsyu, Sikoku and Kyusyu) situates very near to the mainland China, from the viewpoint of spermatopytic distribution. From the charological viewpoint, however, Hondo has the more intimate connection with the Southern-East Adjacent Island, -Ryukyu, Formosa (Taiwan), Philippines and Malay. 2. Courses of migratory birds are shown in the figure. It is presumable that the winter birds have no advantage for the characeous dispersal. On the other hand, summer birds have a possibility to help the dispersal of Charas along their course. 3. Actually, districts of which floras have high value of the affinities of relations with Japan are on the migratory way of those summer birds, or the places where the summer birds spend winter.

View full abstract

The worldwide distribution of Nitellopsis was reviewed here especially based upon the recent works by a few authors. With regards to Japan, the habitats of this plant were first reported by the author, which showed a special type of distribution in Japan, limited in central Honshu, along the Fossa Magna line (Fig. 3). Furthermore, horizontal and vertical distributions in each lake were clarified (Fig. 4-8). That is, the level of Nitellopsis zone in the lake usually consists of the second zone from the deeper limit of the Charophyta plant, and it ranges from 4 to 12 m deep which corresponds to a depth deeper than the depth of transparency in each lake. In Japan, at least in lake Ashino-ko, Nitellopsis seems to be perennial, because the author observed this plant growing luxuriantly there in the winter of 1955 up to 1958.

View full abstract