The Order Palmariales (Rhodophyta) in the North Pacific Area of Russia: Taxonomic Revision of Halosaccion Kützing and Devaleraea Guiry

Research data on taxonomy of the members of the family Palmariaceae (Palmariales, Rhodophyta) from the Russian Pacific coasts are presented. Special attention is paid to taxonomic status of two genera of palmarialean algae inhabiting the area: Halosaccion and Devaleraea . The genus Halosaccion is presented in the Russian Pacific include for sure only one species D. compressa in addition to D. firma. Generic attribution of D. microspora (= H. microsporum ) remains questionable until molecular genetic data are available.


INTRODUCTION
Red palmarialean algae represent one of the most important components of marine algal flora of the Far Eastern seas of Russia [1]. The order Palmariales Guiry et Irvine in Guiry [2] was segregated from the order Rhodymeniales F. Schmitz on the basis of the peculiarities of formation and development of tetrasporangia. The order was considered for a long time to be monotypic with the only family Palmariaceae Guiry [3]. Describing the family, its author (Guiry [3]) included into it the genera Palmaria Stackhouse, Halosaccion Kützing and Leptosarca A. Gepp et E. Gepp. These three genera differed from the other representatives of the order Rhodymeniales sensu stricto by the absence of the carposporophyte generation and the presence of the stalk-cell in the developing tetrasporangia [3]. Later on, the genus Leptosarca was synonymized with the genus Palmaria by Ricker [4]. At present there are 5 genera within the family Palmariaceae according to Schneider and Wynne [5], in addition to already mentioned Palmaria and Halosaccion these are: Neohalosacciocolax I.K. Lee et Kurogi, Devaleraea Guiry and Coriophyllum Setchell et N.L. Gardner in Gardner. In accordance with another viewpoint of Guiry and Guiry [6] there are 4 actual genera in Palmariaceae, whereas the genus Coriophyllum is included in the family Rhodophysemataceae Saunders et McLachlan, [7] which also belongs to the order Palmariales. This family was segregated from the family Acrochaetiaceae Fritsch ex W.R. Taylor by the absence of a carposporophyte and monosporangia, presence of a unique stalk sporangial cell integral to the sexual cycle and abundance of cellular fusions. The Rhodophysemataceae was also differentiated from the Palmariaceae by the presence of Rhodophysema-like tetrasporangia and heteromorphic sexual life history [7]. The ordinal position of the Rhodophysemataceae initially was not definite, however the authors provisionally included it in the Palmariales separating it from the order Acrochaetiales Feldmann [7]. Later on, appropriateness of the inclusion of the Rhodophysemataceae in the order Palmariales was confirmed by molecular data of Saunders et al. [8]. According to modern systematic data, the Rhodophysemataceae includes 5 genera: Coriophyllum, Pseudorhododiscus Masuda, Rhodonematella S.L. Clayden et G.W. Saunders, Rhodophysema Batters and Rhodophysemopsis Masuda [6]. Nearly all of the mentioned genera have a complicated taxonomic history.
The position of the genus Coriophyllum had been uncertain for a long time, but most phycologists referred it to the family Peyssonneliaceae Denizot, within the order Cryptonemiales Schmitz [9]. Now Peyssonneliaceae is considered to belong to a separate order Peyssonneliales D.M. Kayesky, Fredericq et J.N. Norris [10], whereas Coriophyllum is transferred to the family Rhodophysemataceae, order Palmariales [6]. Describing Pseudorhododiscus the author of the taxon (Masuda [11]) had difficulties in assigning it to a definite family and so left it in the category "genera with uncertain taxonomic position". At the present time it is included in the family Rhodophysemataceae [6]. In addition, a relatively recently described genus Rhodonematella S.L. Clayden et G.W. Saunders was placed in the family Rhodophysemataceae [12].
Other members of the Rhodophysemataceae (Rhodophysema, Rhodophysemopsis) formerly were referred to the order Acrochaetiales by Perestenko [13] but this viewpoint was not shared by foreign phycologists. It should be noted that there were two more members considered to belong to the family Rhodophysemataceae -the genera Halosacciocolax S. Lund and Meiodiscus G.W. Saunders et McLachlan [7]. The genus Halosacciocolax had a very complicated taxonomic history. Initially it had been referred to the family Peyssonneliaceae by Lund [14], then to the family Acrochaetiaceae Fritsch ex Taylor by Cabioch and Guiry [15], later on to Palmariaceae by Hawkes and Scagel [16] and at last to Rhodophysemataceae by Saunders and McLachlan [7]. But at present the genus Halosacciocolax has lost its taxonomic independence being merged in synonymy with Rhodophysema by Saunders and Clayden [17], the type genus of the family Rhodophysemataceae.
The genus Meiodiscus was also placed in the family Rhodophysemataceae by the authors of the genus and the family [7,18]. However, later on another family has been described within the order Palmariales -Meiodiscaceae S.L. Clayden [12]. One more family was described within the order Palmariales -Rhodothamniellaceae G.W. Saunders in Saunders et al. [8] that contained two genera -Camontagnea Pujals and Rhodothamniella Feldmann in Christensen [5,6]. Algae of the both genera have uniseriate filamentous structure. In the opinion of Saunders et al. [8] members of this family represent an early genealogic line in the order Palmariales.
Molecular studies of Ragan et al. [19] revealed that members of the order Palmariales had genetic affinity with morphologically distant algae of the orders Acrochaetiales and Nemaliales Schmitz in Engler. In addition to the data on similarity of the pit-plug formation it gave the authors reason to suppose that all three orders had diverged together from main genealogic line of red algae at an early stage of evolution.
According to the present-day systematic data the order Palmariales consists of 4 families: Palmariaceae, Phodophysemataceae, Meiodiscaceae and Rhodothamniellaceae [6].  [20]; the family Meiodiscaceae is presented by Meiodiscus spetsbergensis [13]. The majority of them are small-sized epiphytic or parasitic algae in the form of crusts, cushions or felt. The family Palmariaceae also includes a small parasitic alga Neohalosacciocolax, but the leading role in this family belongs to three genera of mainly epilithic macroalgae: Palmaria, Halosaccion and Devaleraea, widespread on the Russian Pacific coasts [1,21]. The problems of taxonomy of the genus Palmaria were discussed in details in our previous paper [22], the rest 2 genera of palmarialean algae -Halosaccion and Devaleraea -are in the focus of attention of the present study. Main objective of our work is to clarify taxonomic status of these genera and to reconsider generic position of the species referred to them.

MATERIALS AND METHODS
The phycological material used in the present study was collected at the coasts of Eastern Kamchatka from Dezhnev Bay in the north to Lopatka Cape in the south, including Bering Sea, the Commander and Northern Kurile Islands ( Fig. 1) during expeditions of the Hydrobiology Laboratory of Kamchatka Branch of the Pacific Geographical Institute from 1986 to 2015. Algae were collected from May through October on the littoral fringe during low tides, and with use of SCUBA from the depths of 1-5 m according to the standard hydrobiological methods. Algae cast ashore were also sampled. Most part of the studied algae was collected by the author of the present paper manually in Avacha Gulf (south-eastern Kamchatka) and Commander Islands in the intertidal zone during low tides. No special instruments were used. Occasionally other collectors also took part in this work. Fresh material was roughly identified visually in the field, put in plastic bags with clean sea water, labeled and brought either to the laboratory or on board a ship. In case of need the material was kept for some time in refrigerator. Then major part of samples was pressed on herbarium paper, dried, labeled and put for storage in the Herbarium of Kamchatka Branch of the Pacific Geographical Institute (Petropavlovsk-Kamchatskii, Russia). The processing of collections and identification of algae was conducted at the same Institute. The material was sectioned freehand with a razor blade, placed in a drop of fresh water on the slides and examined using light microscopy. The sections were studied unstained. Photos of samples studied were made using photocameras Olympus µ-5010, Olympus SZ-20, Panasonic FS62 Lumix and digital camera for microscope DCM 130.

RESULTS
On the total 77 herbarium sheets containing samples of Halosaccion and 58 sheets with samples of Devaleraea were examined ( Table 1).
Examination of the samples of palmarialean algae from our collections revealed two species of Halosaccion to grow in the studied area -H. glandiforme and H. minjaii. As it follows from the Table, H. glandiforme is the most widespread species in the studied areas. It has wide geographic range and continuous distribution in contrast to H. minjaii which is relatively rare as compared to H. glandiforme and grows mostly on the islands (Commander, Kurile Islands) [31], being only occasionally met on the continental part of Kamchatka [32]. Nevertheless H. minjaii can form dense thickets in associations with coralline algae in the littoral pools and sometimes even compete with H. glandiforme [31].
The genus Devaleraea is presented by three species. D. firma is distributed mostly in the southern part of the studied area where it is rather abundant and sometimes forms monodominant communities on the rocks and stones in the low intertidal zone. D. firma is also in competitive relations with H. glandiforme [33]. The rest two species of the genus Devaleraea -D. compressa and D. microspora have disjunctive areas and are met in the northern part of the studied area. They are common species of the flora however not so abundant as compared to D. firma.
In most cases the members of the genera Halosaccion and Devaleraea are well distinguished even in the field. If plants are membranaceous or thin-coriaceous, saccate, single or aggregated, unbranched, reddish purple to yellowish brown, water filled if undamagedthey most likely belong to the genus Halosaccion. If thalli of the plants are coriaceous, reddish purple to coral red, flattened, sometimes branched -they probably represent the genus Devaleraea. For more detailed identification it is necessary to examine plants anatomy.
A key to genera based on anatomical features is presented below: 1. Large-celled multiseriate cortex with lateral anastomoses and small-celled medulla with stellate protoplast -Halosaccion. 2. Small-celled cortex without anastomoses and large-celled medulla without stellate protoplasts -Devaleraea.
Note: This anatomical approach is reliable in most cases but is not universal. One more known species of the genus Halosaccion -H. yendoi I. K. Lee needs further taxonomic consideration.
Although it was not examined within the frames of the present work because H. yendoi does not grow in the studied areas, its taxonomic status is discussed to some extent. Described originally as H. yendoi [24] it was later transferred to Devaleraea as D. yendoi (Lee) Guiry [ [22]. It seems that Palmaria is the only genus of the three genera of macroalgae of the family Palmariaceae that is relatively easy for identification at the generic level.
Taxonomy of the genus Halosaccion is not so definite. Palmarialean algae of tubular or saccate form with an inner cavity had been attributed to this genus for a long time. Then M.D. Guiry [25] on the basis of vegetative anatomy of the species of Halosaccion showed that they should be divided into two groups. Members of the first group have relatively large-cells multiseriate cortex with lateral anastomoses and relatively small-celled medulla with cytoplasmic outgrowths giving a stellate form to the protoplast. These algae according to Guiry [25] represented the genus Halosaccion sensu stricto. Members of the second group have anatomical features close to Palmaria -small-celled cortex without anastomoses well distinguished from the medulla which is large-celled and without stellate protoplasts. The algae from the second group were referred to a separate genus Devaleraea [25]. Devaleraea ramentacea (Linnaeus) Guiry The decision to transfer H. firmum to the genus Devaleraea is based on the following features of its anatomy: the presence of 2-3-layered largecelled medulla with the protoplasts adjoining to the cell-walls, absence of stellate protoplasts, abrupt transition from the medulla to the smallcelled cortex consisting of 2-3 rows (Fig. 2 C, D) inherent in Devaleraea.

A -algae growing in the intertidal zone; B -fresh samples before pressing on herbarium paper, scale bar = 3 cm; C, D -cross sections of the middle portions of the frond at different magnification power, scale bar = 100 µm
Rightfulness of the attribution of H. firmum to Devaleraea besides that is supported by preliminary data of molecular-genetic analysis of palmarialean algae from Kamchatka [34]. As it followed from the genetic tree constructed using neighbor-joining method on the basis of ribosomal small subunit DNA sequences, algae referred to H. firmum formed common cluster with the species of the genera Palmaria and Devaleraea and differ from other species of All species of the genus Palmaria, in spite of considerable morphological differences, have relatively uniform anatomy: solid blade without cavity, large-celled medulla without stellate protoplasts and small-celled cortex without anastomoses.
In spite of some unresolved taxonomic problems the systematics of the genus Palmaria as a whole is more or less defined [22]. This is not the case with two other genera of the family Palmariaceae -Halosaccion and Devaleraea. Note. Actually there is one more species of Halosaccion known from the Russian Pacific -H. tilesii Kjellman that up to now is treated as an independent one ("current" as it is cited in Algaebase [6]). However L.P. Perestenko [13,26] asserted that its description was based on the material representing in fact Palmaria stenogona that often forms a sort of a hollow cavity in the frond filled with water. This phenomenon was observed repeatedly by her on Commander Islands. So she concluded that the species called Halosaccion tilesii did not really exist [26].
In her later work, Perestenko [27] nevertheless recognized the genus Devaleraea, but reduced its volume to the only species D. ramentacea. Some Japanese phycologists also did not recognize Devaleraea [36,37], and continued to cite D. ramentacea and D. yendoi as the species of the genus Halosaccion. Taxonomic status of the latter species H. yendoi (= D. yendoi) in fact remains uncertain up to now. In the Algaebase There is no certainty also with the type species of the genus Halosaccion. Neolectotypifcation and designation of H. firmum as a type species [25] later on was considered to be erroneous [27].
The problem of lectotypification of the genus Halosaccion is discussed in detail by L.P. Perestenko [27] Perestenko [27] believed that there was no reason in neolectypification of the genus Halosaccion and selection of Н. firmum as a type species. In her opinion the role of the type species of the genus Halosaccion should belong as before to Н. hydrophorum (as it was designated by F. Schmitz [38]. Thus according to Perestenko [27]  In any case her proposal obviously contradicts the law of priority. It is expedient to retain widely used for more than 100 years name -H. glandiforme, and to treat H. hydrophorum as its synonym, in spite of the loss of the type specimen of Ulva glandiformis. In this case H. glandiforme (Fig. 3) should be recognized as the type species of the genus Halosaccion.
So I suggest lectotypification of the genus Halosaccion as follows: On the other hand, both H. glandiforme and H. hydrophorum are now recognized as separate ('current') species (see: Algaebase [6]). Obviously additional studies are necessary to solve these problems.   [31]. Another Russian phycologist Perestenko [27] did not support this transfer as well as validity of division of the genus Halosaccion into two genera. The following arguments were presented by this author to confirm her position: Quite often protoplast is nonstellate in the cells lining the cavity, and it can be non-stellate or have short extensions in the medulla of the membranaceous fronds. "Asterism" can be expressed to a different extent due to the different length of protoplast extensions. On the border between cortex and medulla cells of the inner cortex can fuse and in this case the protoplast acquires laciniate form. It should be noted that Perestenko [27] unlike Guiry [25] distinguished stellate and laciniate protoplasts. She supposed that the extensions of a stellate protoplast were formed as a result of thickening of membrane (cell-wall) or other reasons, except cell fusion, whereas a laciniate protoplast was formed only due to cell fusion. So in her opinion Guiry [25] illustrating Н. americanum I.K. Lee  H. hydrophorum (=H. glandiforme) though they probably have common origin. But our long-term studies showed that these 2 species are separated geographically. Within the Russian Pacific sector H. yendoi is widespread in the Sea of Japan, in the south of the Sea of Okhotsk, on Sakhalin and southern Kurile Islands [31] while H. glandiforme is spread in the northern areas: At the coasts of the Eastern Kamchatka and Commander Islands [21]. According to Perestenko [27] the areas of the two species are overlapping only on Kurile Islands, on the rest of the water area they are allopatric species. So on the basis of genetic data (though preliminary) the status of D. yendoi should be reconsidered and the taxon should be reinstated within the genus Halosaccion, in spite of its anatomy typical of Devaleraea. 3. In contrast to the above said, genetic data on Devaleraea compressa (Fig. 4) obviously support the viewpoint of Klochkova and Selivanova [31]. This species was shown to have close genetic relations with the complex Palmaria-Devaleraea and is considerably different from the members of the genus Halosaccion [34].

Fig. 4. Habit and anatomy of Devaleraea compressa (Ruprecht) Selivanova et Kloczcova
A -algae growing in the intertidal zone; B -fresh samples before pressing on herbarium paper, scale bar = 2 cm; C, D -cross sections of the middle portions of the frond at different magnification power, scale bar = 100 µm Unfortunately we had not yet carried out genetic analyses of D. microspora. So until more detailed study is done we leave this species as a taxon with uncertain generic position.

CONCLUSIONS
This study is a part of multi-year floristic studies on the Pacific coasts of Russia. It showed that red palmarialean algae are well presented there by the genera Halosaccion and Devaleraea (Palmariales, Palmariaceae). The taxonomic revision of both genera carried out in this study revealed that the genus Halosaccion includes for sure 2 species: All three species have typical anatomy of the genus Devaleraea. The belonging of the first two species to this genus is also supported by preliminary genetic data. Generic position of D. microspora (=H. microsporum) is not so obvious because of the absence of genetic data.