College of Architectural Engineering, Weifang University, Weifang 261061, China
2.
Key Laboratory of Submarine Geosciences, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China
3.
Research Center of Continental Dynamics, College of Earth Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China
Funds:
The Resources and Environment Projects of China Ocean Mineral R&D Association under contract No. DY135-E2-2-01; the Natural Science Foundation of Shandong Province under contract No. ZR2020QD076.
The compositional variability of ocean island basalts (OIBs) is thought to reflect partial melting of a lithologically-heterogeneous mantle source dominated by either pyroxenite or peridotite. The Pohnpei Island in Micronesia, which is associated with the Caroline hotspot, is suggested to have been generated from partial melting of a pyroxenite-rich mantle. To examine this hypothesis, we present new major- and trace-element compositions of olivine phenocrysts in basalts from the island. The olivines exhibit large systematic inter- and intra-crystalline compositional variability. In Sample DS1, olivines record compositional zonation, in which cores have relatively high Fo (77–85), Ni (550×10−6–2 392×10−6), and Fe/Mn ratios (66–82), whereas rims have lower Fo (71–78), Ni (526×10−6–1 537×10−6), and Fe/Mn ratios (51–62). By contrast, olivines within other samples preserve no clear compositional zonation, exhibiting similar or slightly lower Fo values (66–78), Ni contents (401×10−6–1 268×10−6), and Fe/Mn ratios (53–69) as the rims of zoned crystals. The distinct chemical contrast between the two different types of olivine suggests they formed in magma chambers at different depths. Analysis using forward petrological modeling and multi-element indicators (Fe/Mn, Zn/Fe, FC3MS (FeOT/CaO−(3×MgO/SiO2)), Mn/Zn, and Ni/(Mg/Fe)) of whole-rock samples and high-Fo olivines is inconsistent with a pyroxenite-rich mantle source. We suggest these inconsistencies reflect an influence on the partition coefficients of Ni and Mn between olivine and liquid during melting at variable pressures and temperatures. In addition, magma recharge and mixing within the magmatic plumbing system can change the composition of olivine. We suggest that identification of the mantle source of OIBs in volcanic islands such as the Pohnpei Island using olivine geochemistry should be treated with caution.
Sponges are an important structure-forming component of the deep-sea benthic fauna, providing biogenic microhabitats to associated organisms and thereby increasing local habitat heterogeneity and supporting high biodiversity, especially when they occur in sponge aggregations (Bell, 2008; Hajdu et al., 2017). Due to the highly complex and exquisite frameworks, the functional significance of euplectellid sponges, well-known for the “Venus flower basket” form and its associated spongicolid shrimps, has been highlighted in enhancing biodiversity and abundance of associated organisms (Kou et al., 2018). Generally, the sponges of family Euplectellidae Gray, 1867 are tubular, cup-like or fungus-like, lophophytose, basiphytose or rhizophytose, and sometimes pedunculate (Shen et al., 2019; Tabachnick, 2002). Megascleres consist of dermal and atrial hexactins (sometimes pentactins or both), choanosomal stauractins, tauactins and diactins (sometimes as well as hexactins and pentactins), and basal anchorate spicules (rarely diactins). Microscleres vary in the outer ends including graphiocomes, floricomes, drepanocomes, and others. There are three subfamilies in this family: Euplectellinae Gray, 1867, Corbitellinae Gray, 1872, and Bolosominae Tabachnick, 2002 (but note that these taxa are likely not monophyletic; see Dohrmann et al., 2017; Dohrmann, 2019; Shen et al., 2019). The subfamily Corbitellinae, in which the typical “Venus flower basket” form is widespread, has 11 genera (4 of them are monotypic) and 39 valid species currently recognized (de Voogd et al., 2021).
The genus Walteria Schulze, 1886 currently consists of two species, W. flemmingii Schulze, 1886 and W. leuckarti Ijima, 1896. They occur in the western Pacific Ocean and eastern Indian Ocean at depths of 370–4 732 m (MacIntosh et al., 2018; Reiswig and Kelly, 2018; Tabachnick, 2002; Tabachnick et al., 2019). The type species W. flemmingii is known from the North Central Pacific Ocean off Hawaii, through the southwest Pacific islands of New Caledonia and New Zealand, extending south to Macquarie Ridge and west to the eastern Indian Ocean of the Perth Canyon in West Australia (MacIntosh et al., 2018; Reiswig and Kelly, 2018; Tabachnick, 2002; Tabachnick et al., 2019). Walteria leuckarti is known from the Sagami Bay, Japan, and Marianas Basin to the southwest Pacific islands around New Zealand (Reiswig and Kelly, 2018). Another indeterminate species of Walteria is known from off Hawaii and New Zealand from depths of 1 446–2 210 m (Reiswig and Kelly, 2018).
By taking W. flemmingii as type species, Schulze (1887) and Tabachnick (2002) described the general morphology and spiculation. Schulze (1887) did not recognize the lateral processes or onychohexaster microscleres, which were supplemented by Tabachnick (2002). However, Tabachnick (2002) made some statements on the existence of prostalia lateralia, large hexactins and pentactins to support the lateral processes, and discasters as some of the disc-bearing microscleres, which were refuted by Reiswig and Kelly (2018) after further collection and examination. Moreover, the existence of primary large stauractins and their importance in generating the rectangular lattice of wall strands were ignored by earlier descriptions, but later underlined by Reiswig and Kelly (2018) in detail. In addition, it was suggested that there were only graphiocomes and floricomes as stellate microscleres (Reiswig and Kelly, 2018). The intraspecific differences between the South Australian specimens having oxyhexasters and New Zealand and Hawaiian specimens having onychohexasters were illustrated (Reiswig and Kelly, 2018). For the valid species, W. leuckarti, which has a unique bottle-brush-like body form, Reiswig and Kelly (2018) indicated that the floricomes are rare and difficult to find; these authors therefore assigned the subspecies W. leuckarti longina Tabachnick, 1988 as a synonymised name so as to avoid subspecies designations within the Hexactinellida. Furthermore, another morphotype, with tubular body, extensive side processes, pointed apex and lattice-like body wall, was suggested to be a new species (Reiswig and Kelly, 2018). This morphotype was previously recorded from off Hawaii and displayed online by the Hawaiian Undersea Research Laboratory (HURL) Archive (http://www.soest.hawaii.edu/HURL/HURLarchive/guide.php). One specimen of dead, eroded framework and lack of loose spiculation (Walteria sp. indet.), coupled with seafloor images presumably identical to it but without collections, from New Zealand was described but not named (Reiswig and Kelly, 2018).
In this study, we describe a new species, the specimens of which were collected by remotely operated vehicle (ROV) at depths of 1 271–1 703 m from the northwestern Pacific seamounts. Local aggregations of this new sponge species and its associated organisms are reported from the Suda Seamount at depths of 1 295–1 320 m. Finally, we present molecular phylogenetic results supporting the placement of the new species in the family Euplectellidae, genus Walteria.
2.
Materials and methods
2.1
Sample and video data collection
Two specimens of Walteria demeterae sp. nov. were collected by ROV from the northwestern Pacific seamounts. The holotype (catalogue number: SIO-POR-244) was collected on September 20, 2018 by ROV Haima from the Weijia Seamount in the northwestern Pacific Ocean at dive ROV12 of cruise DY51/I (12.730 9°N, 156.539 7°E, depth of 1 703 m). The paratype (catalogue number: SIO-POR-099) was collected by ROV Hailong III from the Suda Seamount in the northwestern Pacific Ocean on August 19, 2018 at dive ROV07 of cruise DY48/I (22.010 0°N, 159.377 2°E, depth of 1 271 m). The specimens were preserved in 95% ethanol and deposited in the Sample Repository of Second Institute of Oceanography (SRSIO), Hangzhou, China. A benthic investigation at the Suda Seamount implemented by ROV Hailong III on August 20, 2018 at dive ROV08 of cruise DY48/I (starting point: 22.011 0°N, 159.389 4°E, 1 295 m deep; ending point: 22.012 0°N, 159.391 1°E, 1 307 m deep) recorded a frequent occurrence of a morphotype identical to the new sponge species. Video recordings and still images collected by this investigation, in total 3.0 h, from depths of 1 295–1 320 m, covering 550 m in distance, were examined to describe these local sponge aggregations and the presence of associated macrofauna.
2.2
Spicule and skeleton analysis
Spicules and frameworks were isolated by digesting small pieces of tissue from body wall and lateral processes, which were placed in different centrifuge tubes, in concentrated nitric acid using a metal bath maintained at about 80°C for one day until the tissues were completely dissolved. The spicules and frameworks were then rinsed twice in distilled water and later suspended in alcohol for light microscope (LM) and scanning electron microscope (SEM) observation. Slides for LM were examined with ZEISS Imager A2. Spicules for SEM (TM-1 000 Tabletop Microscope) were pipetted from the solution onto silicon wafers, and skeletons were moved onto silicon wafers, all of which were subsequently coated with platinum. Photomicrographs were obtained by SEM, and measurements were made with the software ZEN pro 2012 1.1.2.0.
2.3
Molecular work
DNA was extracted from small pieces of tissue using a DNeasy Blood&Tissue Kit (Qiagen) according to the manufacturer’s instructions. The DNA quality and concentration were detected with a Nanodrop 2000 and Qubit fluorometer. Then the whole genome was sequenced via the Illumina PE150 strategy on the HiSeqTM X-Ten platform (Illumina, CA, USA) at Zhejiang Tianke High Technology Development Co., Ltd. The total of clean reads was assembled with the SPAdes genome assembler (Bankevich et al., 2012). The nuclear 18S rDNA and 28S rDNA, and mitochondrial 16S rDNA and cytochrome c oxidase subunit I (COI) genes were retrieved from the assembled contigs and used for phylogenetic analysis. The start and end of target sequences were checked by the reported universal PCR primer pairs (Dohrmann et al., 2008; Folmer et al., 1994). The 18S rDNA, 28S rDNA, 16S rDNA, and COI gene sequences are deposited in GenBank with accession numbers of MW652662, MW652660, MW652655, and MW517848, respectively.
2.4
Phylogenetic analysis
The four sequences of 18S rDNA, 28S rDNA, 16S rDNA, and COI plus reference data, were aligned using the ClustalW algorithm with default settings (15/6.66 as gap/gap length penalties) in Geneious prime (Biomatters Ltd., New Zealand). The alignments were performed separately, and then concatenated into one file after manually trimming unalignable regions on both flanks. The phylogenetic tree was inferred with maximum likelihood (ML) analysis using IQ-TREE (Nguyen et al., 2015; Chernomor et al., 2016). The auto-selected best substitution model (-m MFP option) was used in IQ-TREE, with 1 000 bootstrap replicates (command line and input parameters: iqtree -s concatenated_alignments.phy -m MFP -p concatenated_alignments_partition.txt -bb 1 000 -alrt 1 000). The ultrafast bootstrap (UFBoot) was used to assess clade support by adding “-bb 1 000” (Hoang et al., 2018; Minh et al., 2013). An SH-aLRT test (Guindon et al., 2010) was also performed by adding “-alrt 1 000”. The tree was constructed based on the concatenated alignment file of Walteria demeterae sp. nov. and other 31 reference species of hexactinellids by taking three species of the genus Tretopleura as outgroups (detailed information of sequences is shown in Table 1). The ML tree was then visualized and edited in Figtree 1.4.2 (Rambaut, 2006).
Table
1.
GenBank accession numbers of marker genes of the 32 species used in the molecular phylogenetic analysis
Order: family
Species
Voucher number
Accession number
18S rDNA
28S rDNA
16S rDNA
COI gene
Lyssacinosida: Euplectellidae
Walteria demeterae sp. nov. (holotype)
SIO-POR-244
MW652662
MW652660
MW652655
MW517848
Walteria leuckarti
SMF 10522
AM886399
AM886373
AM886337
FR848939
Docosaccus maculatus
GW5429
FM946116
FM946115
FM946105
FR848934
Rhabdopectella tintinnus
HBOI 4-X-88-2-014
AM886402
AM886371
AM886332
FR848941
Regadrella sp.
HBFH 8-VIII-09-2-001
/
FR848916
FR848917
/
Rhizophyta yapensis
SIO-POR-083
MK463603
MK463607
MK458682
MK453399
Acoelocalyx brucei
SMF 10530
AM886401
AM886370
AM886333
FR848938
Malacosaccus coatsi
SMF 10521
AM886400
AM886369
AM886334
FR848937
Euplectella sp. 1
HBOI 19-XI-86-1-001
AM886397
AM886368
AM886335
FR848935
Euplectella sp. 2
HBOI 12-XI-86-1-054
AM886398
AM886367
AM886336
/
Saccocalyx microhexactin
MBM179994
/
/
KM881702
/
Saccocalyx pedunculatus
SMF 12082
MF740862
MF684009
MF683987
/
Hertwigia sp.
USNM 1122181
FM946121
FM946120
FM946104
FR848940
Bolosoma sp.
USNM 1097546
FM946118
FM946117
FM946102
FR848942
Bolosominae n. gen. n. sp.
HURL P4-224-sp7
/
LT627534
LT627520
LT627552
Atlantisella sp.
HBOI 22-X-95-1-7
LT627547
LT627533
LT627519
/
Holascus euonyx
SMF12092
MF740859
/
MF683980
/
Holascus taraxacum
SMF12059
/
MF684005
MF683982
/
Hyalostylus microfloricomus
SMF12085
MF740860
MF684007
MF683984
/
Hyalostylus schulzei
SMF11707
/
/
MF683985
/
Lyssacinosida: Rossellidae
Rossella nuda
SMF 10531
AM886384
AM886355
AM886343
HE580217*
Nodastrella asconemaoida
ZMA POR18484
AM886386
AM886354
AM886344
FR848921
Caulophacus arcticus
SMF 10520
AM886395
AM886360
AM886350
FR819684
Caulophacella tenuis
SMF 10533
AM886392
AM886363
AM886351
FR848927
Caulophacus valdiviae
SMF 10528
AM886394
AM886362
AM886348
FR848929
Caulophacus weddelli
SMF 10527
AM886393
AM886361
AM886349
FR848928
Crateromorpha meyeri
SMF 10525
AM886389
AM886359
AM886347
FR848923
Bathydorus spinosus
SMF 10526
AM886390
AM886358
AM886341
FR848924
Lyssacinosida: Aulocalycidae
Euryplegma auriculare
NIWA 43457
/
LT627535
LT627518
LT627551
Sceptrulophora: Uncinateridae
Tretopleura weijica
SIO-POR-090
/
/
MT176124
MT178277
Tretopleura sp. 1
HURL P4-229-sp10
/
LT627542
LT627529
LT627556
Tretopleura sp. 2
HURL P5-701-sp4
/
LT627543
LT627530
LT627555
Note: The symbol “*” refers that the COI gene of Rossella nuda was from the voucher SMF 11730.
Figure
1.Walteria demeterae sp. nov. specimens. Holotype: in situ images (a1), whole collected specimen including five segments (a2), close-up images of a slightly damaged apex (a3), middle body (a4), and basal part (a5); paratype: in situ image (b1), whole collected specimen including two segments (b2), and close-up image of middle body showing fused skeleton (b3).
Figure
2.Walteria demeterae sp. nov., seafloor images of the same morphotype observed at dive ROV08 of cruise DY48/I from the Suda Seamount in the northwestern Pacific Ocean. a. Living individuals showing different shapes of tubular body with close-up images of small terminal osculum in the pointed apex, circular strands in the inner wall, and lateral processes projecting from the body wall; b. dead individuals showing different degrees of erosion with close-up images of pointed apex and basal part; c. concurrence of both living and dead individuals in one image; d. examples of invertebrates associated with living or dead individuals, including Comatulida, Spongicolidae, Ophioplinthaca defensor, Paguridae, Cirripedia and others (from top to bottom, then from left to right).
Figure
3.Walteria demeteraesp. nov., surfaces of holotype. Dermal surface of body wall with lateral oscula (a1), close-up image of crossing longitudinal and transverse bundles (a2), and SEM of longitudinal bundles (a3); atrial surface of body wall with lateral oscula (b1), close-up image of crossing longitudinal and transverse bundles (b2), and SEM of transverse bundles (b3); lateral processes of dichotomously branching (c1) or simple (c2) outgrowths, the SEM of framework at junction of body wall and lateral process (c3) and its fused diactins (c4); attached stauractins (d1, d2), pentactins (d3, d4) and occasionally hexactins (d5).
Figure
4.Walteria demeterae sp. nov., spicules. a. Choanosomal diactin, whole and enlargements of three types of rays and middle segments; b. choanosomal stauractin, whole and enlargements of axial centre and two types of rays; c. dermal sword hexactin, whole and enlargements of tangential, proximal and two types of distal rays; d. atrial pentactin, whole and enlargements of axial centre, proximal ray, and two types of tangential rays with a smooth or slightly rough tip; e. oxyhexactins, which are sparsely covered by fine spines (located all over the body, e1) or entirely covered by macrospines (restricted to the basal part, e2), whole and enlargements of centres and rays; f. graphiocome, primary centre and terminal ray at same scale, enlargements of primary centre and both the basal part and tip of terminal ray; g. discohexasters in common dense form or rare sparse form (corresponding to “discohexaster 1” and “discohexaster 2” described for W. flemmingii and W. leuckarti by Reiswig and Kelly, 2018), whole and enlargements of terminal ray and disc; h. onychohexaster, whole and enlargements of six types of onychoid tips with 1–5 claws; i. floricome, whole (not intact) and enlargements of terminal ray and tip.
Figure
5.
Maximum-likelihood phylogenetic tree of 32 species from four families of Class Hexactinellida (by taking three species of the genus Tretopleuraas as outgroups), showing the relationships of the species Walteria demeterae sp. nov. and related taxa. The tree was inferred from the concatenated alignments of 18S rDNA, 28S rDNA, 16S rDNA and COI. Numbers at nodes are SH-aLRT/ UFBoot support values (all based on 1 000 replicates).
Material examined. Holotype (Fig. 1a): SIO-POR-244, SRSIO,Haima ROV DY51/I, dive ROV12, collected by Bo Lu, September 20, 2018, at Weijia Seamount in the northwestern Pacific Ocean (12.730 9°N, 156.539 7°E), depth of 1 703 m, stored in 95% ethanol. Paratype (Fig. 1b): SIO-POR-099, SRSIO, Hailong III ROV DY48/I, dive ROV07, collected by Dongsheng Zhang, August 19, 2018, at Suda Seamount in the northwestern Pacific Ocean (22.010 0°N, 159.377 2°E), depth of 1 271 m, stored in 95% ethanol.
Other locations (images only; Fig. 2). Suda Seamount in the northwestern Pacific Ocean: Hailong III ROV DY48/I, dive ROV08, 22.011 0°–22.012 0°N, 159.389 4°–159.391 1°E, depths of 1 295–1 320 m, August 20, 2018.
Description. The new species is presented by a tubular body with lateral oscula and processes (Fig. 1). The holotype has a tubular body with extensive side processes, pointed apex, lattice-like body wall and basiphytous attachment (Fig. 1a). Wall is thin and rigid with numerous oval lateral oscula in irregular arrangement. Distal lateral processes are simple or dichotomously branching outgrowths of the lateral wall. Color of the specimen in ethanol is transparent white.
Five segments are totally 919.8 mm in length and 46.1 mm in maximum diameter (Fig. 1a2). The pointed apex tapers to a very small terminal osculum but is damaged at the top (Fig. 1a3). A large number of lateral processes project about 85 mm from the nodes of the lateral outer surface, forming a crudely hairy aspect (Fig. 1a4). The numerous oval lateral oscula are 15 mm in maximum diameter. Spacing between lateral oscula is nearly adjacent or with maximum edge-to-edge distance of 25 mm. The basal end forms an attachment disc with a smaller diameter than the tubular body attached to hard substrate and supported by lateral struts (Fig. 1a5). The cylindrical body is slightly oblique, the external wall is relatively smooth while the internal wall shows numerous bulging circular strands. The lateral oscula are generally separated by these strands.
The paratype was collected as a long dead, eroded framework with broken lateral processes and lack of loose spiculation (Fig. 1b). Two segments of tubular body and pointed apex are totally 674.4 mm in length and 71.2 mm in maximum diameter (Fig. 1b2). The numerous oval lateral oscula are 20 mm in maximum diameter. Spacing between lateral oscula is nearly adjacent or with maximum edge-to-edge distance of 25 mm. The lateral processes were broken to project at most 48 mm. The pointed apex bends and tapers to a very small terminal osculum but is damaged at the top. Longitudinal, circular and oblique strands can be observed within the framework (Fig. 1b3). Color in ethanol is light brown.
Relevant seafloor images clearly present the unique morphotype characterized by a lattice-work, tubular body with numerous lateral processes, forming a crudely hairy aspect for living individuals or a hispid body wall in different densities and lengths for dead individuals (Fig. 2). The tubular body varies in degree of bending and the pointed apex tapers sharply or tenderly.
Skeleton. The principle skeleton mainly consists of fused diactins with sparsely attached small stauractins, pentactins and sometimes hexactins (Fig. 3). The main skeletal framework of the body wall is a dominant network of crossing longitudinal, transverse and oblique bundles of thick and smooth diactins closely fused by synapticula and therein an irregular framework formed by ray-to-ray fusion of relatively thin diactins at different angles in a plane (Figs 3a and b). Longitudinal bundles are always dermal (Fig. 3a3); transverse (circular) bundles are always the most atrial (Fig. 3b3); oblique bundles and the irregular framework are situated between the longitudinal and transverse bundles. Transverse bundles are strongest and apparent to the naked eye in the collected specimen; longitudinal and oblique bundles are also strong, and the former are much thicker near the base. The framework of lateral processes consists of several parallel diactins fused by synapticula (Fig. 3c). The principle skeleton of the body wall, especially close to the spicule bundles, is sparsely covered by attached small and smooth stauractins, pentactins and occasionally hexactins (Fig. 3d).
The ectosomal skeleton of the dermal surface is composed of a layer of mainly sword-shaped hexactins as dermalia with the short distal ray pointing outwards, covering the outer surface of the main body wall and all surfaces of the lateral processes. The inner surface of the body wall is covered by a layer of pentactins as atrialia, with the long proximal rays extending into the wall. Microscleres are dispersed among the proximal rays of dermalia and atrialia.
Spicules. Megascleres of the body are choanosomal diactins and stauractins, dermal hexactins, and atrial pentactins (Figs 4a-d; Table 2).
Choanosomal diactins (Fig. 4a) vary widely in size with 451.0–(1 071.4)–4 300.0 µm in length (some can be longer) and 8.0–(14.0)–27.5 µm in diameter (Table 2). They are straight or gently curved, subterminally rough or rarely smooth, with a generally inconspicuous or evident central swelling, rounded or conically pointed terminal ends, sometimes slightly inflated near the tips.
Choanosomal stauractins (Fig. 4b) are rarely observed and have lateral rays with rounded or blunt-pointed terminal ends which are rough and slightly inflated near the tips. The lateral rays of stauractins are generally unequal with the long ray 168.5–(265.9)–738.9 µm in length and the short ray 107.5–(157.6)–211.3 µm in length; the ray width is 8.8–(14.2)–18.5 µm (Table 2). There is a nearly undetectable swelling or sometimes a thin and very short fifth ray in the axial cross.
Dermalia (Fig. 4c) of the body wall and lateral processes are similar sword hexactins with proclined spines covering the terminal end of each ray. The distal ray is inflated lanceolate, sometimes oval or occasionally uninflated (or slightly inflated), 88.9–(172.0)–275.0 µm long in the lateral processes, longer than that in the body wall with 49.2–(136.2)–215.0 µm in length; tangential rays have rounded and often slightly inflated terminal ends, with a mean length of about 122 µm both in the body wall and lateral processes but varying more widely in size in the body wall than in the lateral processes; the proximal ray is 1.5–(4.2)–9.4 times longer than the tangential rays with conically pointed tips, which is 243.3–(506.6)–882.0 µm long and 11.3–(14.5)–17.3 µm thick in the lateralas processes, larger than that in the body wall with 114.0–(431.9)–686.7 µm in length and 6.4–(13.6)–18.7 µm in width (Table 2). Dermal sword hexactins are more common in the lateral processes but more variable in the body wall.
Atrialia are regular smooth pentactins (Fig. 4d), mainly present within the body wall, with a small or rarely spherical swelling; tangential rays are 111.5–(158.0)–225.0 µm long with tips smooth or slightly rough, rounded or bluntly pointed, rarely inflated; the proximal ray varies in length of 83.3–(310.2)–654.0 µm with slightly rough and blunt-pointed tips (Table 2).
Microscleres are common oxyhexactins and graphiocomes, a number of discohexasters, some onychohexasters, and rare floricomes (Figs 4e-i; Table 2).
Oxyhexactins (Fig. 4e) in the whole body wall and lateral processes are sparsely covered by fine spines over the entire surface or rarely smooth, both similar in size and shape with rougher and parabolic terminal ends (smooth-rough oxyhexactins; Fig. 4e1). In addition, there is another kind of oxyhexactin entirely covered by macrospines, 3.0–(7.1)–12.0 µm in maximum length, with sharply pointed tips (macrospined oxyhexactins; Fig. 4e2; Table 2), which are restricted to the basal part. Rays of the former are 115.0–(180.3)–265.0 µm long and 3.9–(8.0)–12.4 µm thick, larger than the macrospined ones with 71.5–(117.7)–195.3 µm in length and 2.3–(3.0)–4.0 µm in width (Table 2).
Graphiocomes (Fig. 4f) are typical for Euplectellidae with 22.7–(25.6)–29.7 µm in centre diameter; the primary rays are 6.3–(7.5)–10.0 µm long, ornamented with short smooth hemispheric knobs or rarely smooth; terminal rays (raphides) are very thin and 100.0–(158.9)–179.0 µm long, bearing sparse recurved spines (Table 2); each primary ray supports on its distal swelling about 60 bases of raphides.
Discohexasters (Fig. 4g) are spherical, 77.5–(88.9)–110.0 µm in diameter, with very short smooth primary rays 3.5–(5.8)–9.0 µm in length and rough straight terminal rays 30.0–(35.7)–46.1 µm in length (Table 2); they occur in either the more common dense form with more than 12 terminal rays per primary ray or rare sparse form with about 6 terminal rays on each primary ray; marginal teeth on terminal discs number 11–13.
Onychohexasters (Fig. 4h) are 70.0–(83.0)–100.0 µm in diameter, a little smaller than discohexasters; the primary rays are 2.5–(4.1)–6.7 µm long, with a bulge extending beyond the origins of the terminal rays; the terminal rays are 30.0–(37.0)–45.2 µm in length, covered by either fine or large recurved spines (Table 2); onychoid tips in individual spicules are often distinct with 1–5 claws.
Floricomes (Fig. 4i) were only observed thrice in the body wall, with a diameter of 84.7–(122.1)–184.0 µm; the primary rays are smooth and 3.8–(5.8)–8.3 µm long; terminal rays are 33.3–(53.5)–83.3 µm long and at about nine per primary ray with five terminal claws, of which the outer surfaces are sparsely covered by fine spines (Table 2).
Etymology. The new species name demeterae refers to the goddess of fertility and harvest in Greek mythology named Demeter, referring to the abundant organisms often associating with the new species as well as its attractive appearance in situ.
Remarks. The new species described here is tubular, basiphytous but not pedunculate, bearing choanosomal diactins and stauractins, dermal hexactins, atrial pentactins and various microscleres, and is therefore clearly attributed to the subfamily Corbitellinae of family Euplectellidae (Tabachnick, 2002). Its extensive lateral processes, numerous oval lateral oscula which are irregularly situated in the body wall, the presence of microscleres with oxyoidal (microhexactins and graphiocomes), discoidal and onychoidal outer ends, and the absence of anchorate discohexasters, suggest it belongs to the genus Walteria (Tabachnick, 2002), which is also confirmed by our molecular phylogenetic study (see Section 3.3 below).
The new species shares the form of a hollow body with lateral oscula and processes with the two valid species and their loose spiculation is highly similar in composition and form, but the new species is significantly differentiated in the framework of the body wall. Walteria flemmingii has a net-like, barrel-shaped body with a reticulate wall consisting of a sieve-like network of thin strands; the central elements of strands are primarily large stauractins and diactins generally fused by synapticula into a rigid network. The new species has a tubular body with thin wall and pointed apex similar to that of W. flemmingii but is distinguished by its principle skeleton consisting mostly of diactins (without substantial contribution of stauractins). For the loose spiculation, there are choanosomal tauactins, floricomes in variable abundance, and sometimes oxyhexasters occurring in W. flemmingii, but none of choanosomal tauactins or oxyhexasters was observed in the new species and only three floricomes were observed in the holotype.
Walteria leuckarti has a bottle-brush-like body with a slender, tubular main stem composed of diactins soldered together at contact points or connected by synapticula to form a thick stem wall with oval lateral oscula. The new species has primary spicules of the body wall framework as diactins and the oval lateral oscula similar to that of W. leuckarti but is distinguished by its much larger diameters of both cylindrical body and lateral oscula while much smaller thickness of the body wall. For the loose spiculation, there are choanosomal tauactins and sometimes oxyhexasters occurring in W. leuckarti, but none of these two types of spicules was observed in the new species.
In addition, the dead and eroded framework (Walteria sp. indet.) described by Reiswig and Kelly (2018) was a choanosomal skeleton consisting of a framework of fused diactins, suggesting a tubular body with oval lateral oscula but no evidence of lateral processes. Its lateral oscula are much more pronounced than in the paratype of the new species, which was also collected as an eroded framework; neither transverse or circular bundles are obvious, and nor do spicules attach to the longitudinal bundles, suggesting it is different from the new species described here. Furthermore, the related records of seafloor images from off Hawaii and New Zealand are almost identical to the appearance of the new specimens and in situ images from this study and can be allocated to the same morphotype rather than the eroded framework (Walteria sp. indet.) described by Reiswig and Kelly (2018). But whether they are the same species retains to be determined until a related living specimen from off Hawaii or New Zealand is collected.
3.2
Ecology of the new sponge species
Walteria demeterae sp. nov. is distributed in the northwestern Pacific seamounts (depths of 1 271–1 703 m; specimens and images). The bottom disc of the tubular body is attached to hard substrate. Local aggregations of Walteria demeterae sp. nov. were observed on the Suda Seamount in the northwestern Pacific Ocean at depths of 1 295–1 320 m (Fig. 2). At least 56 individuals characterized by the unique morphotype of the new sponge species were counted, representing 90% of the total individuals of sponges recorded in the track covering 550 m in distance. There were 33 individuals alive and others dead in different degrees of erosion.
Numerous ophiuroids (such as Ophioplinthaca defensor, Ophioleila elegans), crinoids (Comatulida), crustaceans (such as Spongicolidae, Paguridae, Cirripedia) and other taxa were often observed associating with both living and dead individuals of Walteria demeterae sp. nov. (Fig. 2). Echinoderms, in particular crinoids and ophiuroids, were often found in high abundances attached to the sponges, with ophiuroids on or inside the canals and several crinoids often on the upper portion (Beazley et al., 2013; Na et al., 2021). Spongicolid shrimps, which often take euplectellid sponges as their hosts (Kou et al., 2018; Xu et al., 2017), were observed on or inside the canals. Other crustaceans, such as hermit crabs and barnacles, were observed to prefer dead individuals of Walteria demeterae sp. nov. over living ones.
3.3
Molecular phylogenetic analysis of the new sponge species
The length of the 18S rDNA, 28S rDNA,16S rDNA, and COI gene fragments was confirmed by respective primers sequence (Dohrmann et al., 2008; Folmer et al., 1994), and the PCR primers of the four target fragments could be located in the correct position of the marker genes. The data set of 18S rDNA, 28S rDNA, 16S rDNA partial sequence and COI complete sequence of Walteria demeterae sp. nov. (holotype; catalogue number: SIO-POR-244) contained 1 955 bp, 1 225 bp, 495 bp and 709 bp, respectively. The phylogeny confirms that the new species is closely related to W. leuckarti, supporting its systematic position within the family Euplectellidae, genus Walteria (Fig. 5).
4.
Conclusions
According to both morphological and molecular assessment, the hexactinellid specimens described here clearly belong to a species new to science and are attributed to the genus Walteria (family Euplectellidae). This new species, Walteria demeterae sp. nov., shows a unique morphotype characterized by a lattice-work, tubular body with numerous lateral processes. Its structural complexity plays a key role in providing biogenic microhabitats and accordingly increasing biodiversity and abundance of associated organisms in the deep sea. Due to the rare specimens but relatively common records in seafloor images of the morphotype of this new sponge species, new materials should be collected and analyzed to examine intraspecific differences.
The method we used here for searching marker genes in genome-level datasets of marine macrobenthos has proven effective when applied to sponges. It can also provide basic information to design or edit primers when amplification of the gene fragments of relevant species is needed.
Acknowledgements
We thank all the scientists and crew on the R/V Haiyang 6 and Dayang Yihao and the ROV Haima and Hailong III teams, for their dedicated research work and collection of deep-sea specimens. We thank Dohrmann Martin for the constructive suggestions.
Ammannati E, Jacob D E, Avanzinelli R, et al. 2016. Low Ni olivine in silica-undersaturated ultrapotassic igneous rocks as evidence for carbonate metasomatism in the mantle. Earth and Planetary Science Letters, 444: 64–74,
[2]
Collinet M, Charlier B, Namur O, et al. 2017. Crystallization history of enriched shergottites from Fe and Mg isotope fractionation in olivine megacrysts. Geochimica Et Cosmochimica Acta, 207: 277–297,
[3]
Coogan L A, Saunders A D, Wilson R N. 2014. Aluminum-in-olivine thermometry of primitive basalts: Evidence of an anomalously hot mantle source for large igneous provinces. Chemical Geology, 368: 1–10,
[4]
Courtillot V, Davaille A, Besse J, et al. 2003. Three distinct types of hotspots in the Earth’s mantle. Earth and Planetary Science Letters, 205(3–4): 295–308,
[5]
De Hoog J C M, Gall L, Cornell D H. 2010. Trace-element geochemistry of mantle olivine and application to mantle petrogenesis and geothermobarometry. Chemical Geology, 270(1–4): 196–215,
[6]
De Maisonneuve C B, Costa F, Huber C, et al. 2016. How do olivines record magmatic events? Insights from major and trace element zoning. Contributions to Mineralogy and Petrology, 171(6): 56. doi: 10.1007/s00410-016-1264-6
[7]
Dixon T H, Batiza R, Futa K, et al. 1984. Petrochemistry, age and isotopic composition of alkali basalts from Ponape Island, western Pacific. Chemical Geology, 43(1–2): 1–28,
[8]
Elkins L J, Bourdon B, Lambart S. 2019. Testing pyroxenite versus peridotite sources for marine basalts using U-series isotopes. Lithos, 332–333: 226–244,
[9]
Foley S F, Prelevic D, Rehfeldt T, et al. 2013. Minor and trace elements in olivines as probes into early igneous and mantle melting processes. Earth and Planetary Science Letters, 363: 181–191,
[10]
Gleeson M L M, Gibson S A. 2019. Crustal controls on apparent mantle pyroxenite signals in ocean-island basalts. Geology, 47(4): 321–324,
[11]
Gómez-Ulla A, Sigmarsson O, Gudfinnsson G H. 2017. Trace element systematics of olivine from historical eruptions of Lanzarote, Canary Islands: Constraints on mantle source and melting mode. Chemical Geology, 449: 99–111,
[12]
Heinonen J S, Fusswinkel T. 2017. High Ni and low Mn/Fe in olivine phenocrysts of the Karoo meimechites do not reflect pyroxenitic mantle sources. Chemical Geology, 467: 134–142,
[13]
Herzberg C. 2006. Petrology and thermal structure of the Hawaiian plume from Mauna Kea volcano. Nature, 444(7119): 605–609,
[14]
Herzberg C. 2011. Identification of source lithology in the Hawaiian and Canary Islands: Implications for origins. Journal of Petrology, 52(1): 113–146,
[15]
Herzberg C, Asimow P D. 2008. Petrology of some oceanic island basalts: PRIMELT2. XLS software for primary magma calculation. Geochemistry, Geophysics, Geosystems, 9(9): Q09001,
[16]
Herzberg C, Asimow P D, Ionov D A, et al. 2013. Nickel and helium evidence for melt above the core–mantle boundary. Nature, 493(7432): 393–397,
[17]
Herzberg C, Cabral R A, Jackson M G, et al. 2014. Phantom Archean crust in Mangaia hotspot lavas and the meaning of heterogeneous mantle. Earth and Planetary Science Letters, 396: 97–106,
[18]
Herzberg C, Vidito C, Starkey N A. 2016. Nickel-cobalt contents of olivine record origins of mantle peridotite and related rocks. American Mineralogist, 101(9): 1952–1966,
[19]
Hirose K, Kushiro I. 1993. Partial melting of dry peridotites at high pressures: Determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters, 114(4): 477–489,
[20]
Hofmann A W, White W M. 1982. Mantle plumes from ancient oceanic crust. Earth and Planetary Science Letters, 57(2): 421–436,
[21]
Howarth G H, Harris C. 2017. Discriminating between pyroxenite and peridotite sources for continental flood basalts (CFB) in southern Africa using olivine chemistry. Earth and Planetary Science Letters, 475: 143–151,
[22]
Huang Shichun, Zheng Yongfei. 2017. Mantle geochemistry: Insights from ocean island basalts. Science China Earth Sciences, 60(11): 1976–2000,
[23]
Jackson M G, Price A A, Blichert-Toft J, et al. 2017. Geochemistry of lavas from the Caroline hotspot, Micronesia: Evidence for primitive and recycled components in the mantle sources of lavas with moderately elevated 3He/4He. Chemical Geology, 455: 385–400,
[24]
Keating B H, Mattey D P, Helsley C E, et al. 1984. Evidence for a hot spot origin of the Caroline Islands. Journal of Geophysical Research: Solid Earth, 89(B12): 9937–9948,
[25]
Kushiro I. 1996. Partial melting of a fertile mantle peridotite at high pressures: An experimental study using aggregates of diamond. In: Basu A, Hart S, eds. Earth Processes: Reading the Isotopic Code. Geophysical Monograph Series. Washington, DC: American Geophysical Union, 95: 109–122,
[26]
Larrea P, França Z, Lago M, et al. 2013. Magmatic processes and the role of antecrysts in the genesis of Corvo Island (Azores Archipelago, Portugal). Journal of Petrology, 54(4): 769–793,
[27]
Le Roux V, Lee C T A, Turner S J. 2010. Zn/Fe systematics in mafic and ultramafic systems: Implications for detecting major element heterogeneities in the Earth’s mantle. Geochimica Et Cosmochimica Acta, 74(9): 2779–2796,
[28]
Lee C T A, Luffi P, Plank T, et al. 2009. Constraints on the depths and temperatures of basaltic magma generation on Earth and other terrestrial planets using new thermobarometers for mafic magmas. Earth and Planetary Science Letters, 279(1–2): 20–33,
[29]
Li C S, Ripley E M. 2010. The relative effects of composition and temperature on olivine-liquid Ni partitioning: Statistical deconvolution and implications for petrologic modeling. Chemical Geology, 275(1–2): 99–104,
[30]
Li Xiaohui, Zeng Zhigang, Dan Wei, et al. 2020. Source lithology and crustal assimilation recorded in low δ18O olivine from Okinawa Trough, back-arc basin. Lithos, 360–361: 105444,
[31]
Lissenberg C J, MacLeod C J. 2016. A reactive porous flow control on mid-ocean ridge magmatic evolution. Journal of Petrology, 57(11–12): 2195–2220,
[32]
Liu Yongsheng, Hu Zhaochu, Gao Shan, et al. 2008. In situ analysis of major and trace elements of anhydrous minerals by LA-ICP-MS without applying an internal standard. Chemical Geology, 257(1–2): 34–43,
[33]
Mattey D P. 1982. The minor and trace element geochemistry of volcanic rocks from Truk, Ponape and Kusaie, eastern Caroline Islands; the evolution of a young hot spot trace across old Pacific Ocean crust. Contributions to Mineralogy and Petrology, 80(1): 1–13,
[34]
Matzen A K, Baker M B, Beckett J R, et al. 2013. The temperature and pressure dependence of nickel partitioning between olivine and silicate melt. Journal of Petrology, 54(12): 2521–2545,
[35]
Matzen A K, Baker M B, Beckett J R, et al. 2017b. The effect of liquid composition on the partitioning of Ni between olivine and silicate melt. Contributions to Mineralogy and Petrology, 172(1): 3. doi: 10.1007/s00410-016-1319-8
[36]
Matzen A K, Wood B J, Baker M B, et al. 2017a. The roles of pyroxenite and peridotite in the mantle sources of oceanic basalts. Nature Geoscience, 10(7): 530–535,
[37]
McDonough W F, Sun S S. 1995. The composition of the Earth. Chemical Geology, 120(3−4): 223–253. doi: 10.1016/0009-2541(94)00140-4
[38]
Morgan W J. 1971. Convection plumes in the lower mantle. Nature, 230(5288): 42–43,
[39]
Niu Yaoling, Wilson M, Humphreys E R, et al. 2011. The origin of intra-plate ocean island basalts (OIB): The lid effect and its geodynamic implications. Journal of Petrology, 52(7–8): 1443–1468,
[40]
O'Neill H S C, Jenner F E. 2016. Causes of the compositional variability among ocean floor basalts. Journal of Petrology, 57(11–12): 2163–2194,
[41]
Paquet M, Cannat M, Brunelli D, et al. 2016. Effect of melt/mantle interactions on MORB chemistry at the easternmost Southwest Indian Ridge (61°–67°E). Geochemistry, Geophysics, Geosystems, 17(11): 4605–4640,
[42]
Putirka K D. 2008. Introduction to minerals, inclusions and volcanic processes. Reviews in Mineralogy and Geochemistry, 69(1): 1–8,
[43]
Putirka K, Tao Y, Hari K R, et al. 2018. The mantle source of thermal plumes: Trace and minor elements in olivine and major oxides of primitive liquids (and why the olivine compositions don't matter). American Mineralogist, 103(8): 1253–1270,
[44]
Rehkämper M, Hofmann A W. 1997. Recycled ocean crust and sediment in Indian Ocean MORB. Earth and Planetary Science Letters, 147(1–4): 93–106,
[45]
Rehman H U, Nakaya H, Kawai K. 2013. Geological origin of the volcanic islands of the Caroline group in the Federated States of Micronesia, western Pacific. South Pacific Studies, 33(2): 101–118
[46]
Ren Zhongyuan, Takahashi E, Orihashi Y, et al. 2004. Petrogenesis of tholeiitic lavas from the submarine Hana ridge, Haleakala volcano, Hawaii. Journal of Petrology, 45(10): 2067–2099,
[47]
Shaikh A M, Patel S C, Bussweiler Y, et al. 2019. Olivine trace element compositions in diamondiferous lamproites from India: Proxies for magma origins and the nature of the lithospheric mantle beneath the Bastar and Dharwar cratons. Lithos, 324–325: 501–518,
[48]
Shorttle O. 2015. Geochemical variability in MORB controlled by concurrent mixing and crystallisation. Earth and Planetary Science Letters, 424: 1–14,
[49]
Sobolev A V, Hofmann A W, Kuzmin D V, et al. 2007. The amount of recycled crust in sources of mantle-derived melts. Science, 316(5823): 412–417
[50]
Sobolev A V, Hofmann A W, Sobolev S V, et al. 2005. An olivine-free mantle source of Hawaiian shield basalts. Nature, 434(7033): 590–597,
[51]
Spengler S R. 1990. Geology and hydrogeology of the island of Pohnpei, Federated States of Micronesia [dissertation]. Honolulu, HI: University of Hawaii at Manoa
[52]
Spengler S, Spencer K, Mahoney J G G. 1994. Geology and geochemical evolution of lavas on the island of Pohnpei, Federated States of Micronesia. http://e2hi.s3.amazonaws.com/element-environmental-llc/uploads/attachment/attachment/32/12_Geology_and_Geochemical_Evolution_of_Lavas_on_the_Island_of_Pohnpei.pdf [2021-01-01]/[2021-04-20]
[53]
Stracke A. 2012. Earth’s heterogeneous mantle: A product of convection-driven interaction between crust and mantle. Chemical Geology, 330–331: 274–299,
[54]
Sun S S, McDonough W F. 1989. Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes. Geological Society London Special Publications, 42(1): 313–345. doi: 10.1144/gsl.sp.1989.042.01.19
[55]
Thompson R N, Gibson S A. 2000. Transient high temperatures in mantle plume heads inferred from magnesian olivines in Phanerozoic picrites. Nature, 407(6803): 502–506,
[56]
Vidito C, Herzberg C, Gazel E, et al. 2013. Lithological structure of the Galápagos plume. Geochemistry, Geophysics, Geosystems, 14(10): 4214–4240,
[57]
Walter M J. 1998. Melting of garnet peridotite and the origin of komatiite and depleted lithosphere. Journal of Petrology, 39(1): 29–60,
[58]
Wang Xuance, Li Zhengxiang, Li Xianhua, et al. 2012. Temperature, pressure, and composition of the mantle source region of late Cenozoic basalts in Hainan Island, SE Asia: A consequence of a young thermal mantle plume close to subduction zones?. Journal of Petrology, 53(1): 177–233,
[59]
Yang Zongfeng, Zhou Junhong. 2013. Can we identify source lithology of basalt?. Scientific Reports, 3(1): 1856. doi: 10.1038/srep01856
[60]
Yaxley G M. 2000. Experimental study of the phase and melting relations of homogeneous basalt + peridotite mixtures and implications for the petrogenesis of flood basalts. Contributions to Mineralogy and Petrology, 139(3): 326–338,
[61]
Yaxley G M, Green D H. 1998. Reactions between eclogite and peridotite: Mantle refertilisation by subduction of oceanic crust. Schweizerische Mineralogische und Petrographische Mitteilungen, 78(2): 243–255
[62]
Zamboni D, Trela J, Gazel E, et al. 2017. New insights into the Aeolian Islands and other arc source compositions from high-precision olivine chemistry. Lithos, 272–273: 185–191,
[63]
Zhang Guoliang, Wang Shuai, Zhang Ji, et al. 2020a. Evidence for the essential role of CO2 in the volcanism of the waning Caroline mantle plume. Geochimica et Cosmochimica Acta, 290: 391–407,
[64]
Zhang Guoliang, Zhang Ji, Wang Shuai, et al. 2020b. Geochemical and chronological constraints on the mantle plume origin of the Caroline Plateau. Chemical Geology, 540: 119566. doi: 10.1016/j.chemgeo.2020.119566
[65]
Zong Tong, Li Zhenggang, Dong Yanhui, et al. 2020. Geochemical constraints on mantle melting and magma genesis at Pohnpei island, Micronesia. Minerals, 10(9): 816. doi: 10.3390/min10090816
Long YUAN, Quanshu YAN. Source Lithology and Magmatic Processes Recorded in the Mineral of Basalts from the Parece Vela Basin. Acta Geologica Sinica - English Edition, 2022. doi:10.1111/1755-6724.14937
Figure 1. Location and altitude map of the Pohnpei Island in the Pacific Ocean from Zong et al. (2020) (a) and map of the Pohnpei Island showing sample locations (b).
Figure 2. Hand specimens (a, b), photomicrographs showing the textural characteristics of olivine (c, d), and back-scattered electron (BSE) (e, f) images of basalts from sample locations DS1 and DS2 on the Pohnpei Island.
Figure 3. CaO (a), Ni (b), MnO (c), and FeOT (d) contents plotted against MgO contents of basalts from the Pohnpei Island. Literature data are from Mattey (1982), Dixon et al. (1984), and Jackson et al. (2017), and Stage 1 and Stage 2 basalts are from Zhang et al. (2020a). a. The broken black line representing the boundary separating peridotite- and pyroxenite-derived primary magmas is from Herzberg and Asimow (2008). The green line represents the liquid line of descent formed by crystallization of olivine, then olivine (Ol) + clinopyroxene (Cpx) + plagioclase (Plg), following Herzberg and Asimow (2008). b. The black line represents the calculated Ni contents for partial melts of fertile peridotite formed by accumulated fractional melting. The insets in b and c followed Herzberg (2011). c. The MnO contents of experimental melts from peridotite KR-4003 (Walter, 1998) are shown in the inset, and the numbers 2 to 7 are pressures (108 Pa).
Figure 4. Variations of Ca (a), Ni (b), Mn (c) contents and Fe/Mn ratios (d) against Fo values of olivines in the Pohnpei basalts. The data of Pohnpei olivines analyzed by Zong et al. (2020) and reported by Zhang et al. (2020a) for Stage 1 and Stage 2 basalts are also shown. Olivine data for MORB and Hawaiian and Icelandic basalts are from Sobolev et al. (2007). Black fields represent the calculated Ca, Ni, Mn, and Fe/Mn ratios in olivines crystallizing from primary melts derived from a fertile peridotite source, and contain 1 964 ×10−6 Ni, 3.45 wt.% CaO, and 0.13 wt.% MnO (Herzberg, 2011), the pink field represents the 1σ uncertainty for Ni (Herzberg et al., 2013). Light blue fields represent calculated olivine compositions from olivine-fractionated derivative liquids with 8–20 wt.% MgO. The assumed primary magma was that produced in experiment 40.06 by Walter (1998) on fertile peridotite KR-4003. Numbered black lines are for olivines modeled from olivine-fractionated derivative liquids with 20–38 wt.% MgO.
Figure 5. FeO/MnO ratios (a), Zn/Fe ratios (b), and FC3MS (FeOT/CaO − (3 × MgO/SiO2)) (c) ratios in olivines from the Pohnpei basalts. Literature data are from Mattey (1982), Dixon et al. (1984), and Jackson et al. (2017), and Stage 1 and Stage 2 basalts are from Zhang et al. (2020a). Inferred primary magmas for whole-rocks with >10 wt.% MgO were calculated using the model of Lee et al. (2009). The typical field for peridotite-derived melts compositions follows Herzberg (2011) in the diagram of FeO/MnO vs MgO, and Le Roux et al. (2010) in the diagram of Zn/Fe vs MgO. The boundary between peridotite-and pyroxenite-derived melts is based on the upper limit values of peridotite melts from Yang and Zhou (2013).
Figure 6. 100 (Mn/Fe) vs Ni/(Mg/Fe) ratios/103 (a), 104 (Zn/Fe) ratios (b), Mn/Zn ratios (c), and Ca/Al ratios (d) in olivines from the Pohnpei basalts. Data for the Pohnpei olivines analyzed by Zong et al. (2020) and the Stage 1 and Stage 2 olivines reported by Zhang et al. (2020a) are also shown. Olivine data from MORB, the WPM-THIN and WPM-THICK basalts, and the fields for olivine crystallized from pyroxenite- and peridotite-derived melts are from Sobolev et al. (2007).
Figure A1. Comparison of MnO (a), FeO (b), Ni (c) contents and Fe/Mn ratios (d) in the analyzed olivines obtained by LA-ICP-MS and EPMA, respectively.
Figure A2. Primitive mantle (PM; McDonough and Sun, 1995)-normalized trace element patterns of Pohnpei lavas. OIBs (ocean island basalts) trace element data are from Sun and McDonough (1989).
Figure A3. Representative photomicrographs and analytical points with chemical compositions for representative olivine crystal in DS1. The yellow points were the analytical spots by electron probe microanalyser (EPMA).
Figure A4. The representative trace elements ratios against Fo values of the olivines from Pohnpei basalts. The data of Pohnpei olivines analyzed by Zong et al. (2020) and Pohnpei Stage 1 and Stage 2 basalts reported by Zhang et al. (2020a) are also plotted. The data of olivines in MORBs, Hawaii and Iceland basalts are from Sobolev et al. (2007).
DownLoad:
DownLoad:
DownLoad:
