Genetic variability in microbial eukaryotes reshapes marine biodiversity assessment in the age of amplicon sequencing

These cases point to examples in which solitary and colonial Collodaria shared the same dominant sequence variant, or in the case of coll-012, a closely related ASV. Such examples are evidence that solitary and colonial specimens, previously described as different morphospecies, may actually repre- sent different life stages of the same taxon. A full version of this phylogenetic tree that depicts all specimens’ dominant ASVs can be found in S4 Fig. Sequence data underlying this tree can be found in S1 Data and S1 Table, and tree parameters are detailed in “Materials and methods”. https://doi.org/10.1371/journal.pone.0326053.g008

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 18 / 29 Another key finding from the ASV composition of these juvenile radiolarians is that they suggest no change in intrag- enomic richness occurs during ontogeny. The juvenile Acanthodesmidae specimen yielded 2 ASVs, which is within the range of ASV richness found in mature specimens morphologically assigned to the same family (Acanthodesmida- vinculata-01, Acanthodesmida-vinculata-02, and acanthodesmid-sp-01; 1, 2, and 4 ASVs, respectively; Fig 5b). Simi- larly, the mature Spumellaria specimen (classified as lithellid sp. 3) yielded 40 ASVs, while the two juveniles putatively belonging to the same taxon yielded 39 and 22 ASVs each. Thus, mature specimens of both Nassellaria and Spumel- laria appear to have intragenomic variation that is consistent with juvenile individuals, making it improbable that radio- larians accumulate intragenomic variation throughout their life cycles, as ciliates do [59]. This result means that juvenile specimens may be successfully classified based on their ASV composition in precisely the same way as fully-mature specimens in metabarcoding surveys. Metabarcoding surveys therefore have the potential to reveal a more complete picture of radiolarian community composition than microscopy counts or in-situ imaging due to their inherent inclusion of juvenile specimens that are too indistinct to be morphologically identified (see S5 Fig). Our results also suggest that Collodaria do not accumulate genetic diversity during cell division and colony growth,

and different life stages yield comparable genetic signatures. Collodaria colonies have similar magnitudes of intrag- enomic richness as the individual cells which comprise them, as well as specimens with a solitary life habit. Sipho- nosphaera socialis was sampled as whole colonies of 10s–100s of cells (3 specimens) and also as individual cells picked from different colonies (3 specimens). Single-cells yielded a mean of 12.3 ASVs (sd = 0.6), whereas entire colonies yielded fewer ASVs on average (mean = 6) but more variation in ASV richness (sd = 7). Likewise, Collodaria specimens with solitary versus colonial life habits did not show consistently different magnitudes of intragenomic richness. Solitary collodarians yielded a mean of 35 ASVs each, whereas colonial specimens yielded between 16 and 50 ASVs on average (depending on whether the exceptionally-diverse Collosphaera huxleyi was excluded or included, respectively) (Fig 4; S1 Data). We interpret this as support for the hypothesis that Collodaria grow their colonies via binary fission and are made up of individuals that are genetically identical to the originating cell (e.g., [60]). Thus, the genetic richness observed in Collodaria does not appear to be directly related to the number of cells making up their colonies. Sequence information also has importance for understanding the roles of naked (un-skeletonized) solitary morpho- types, which may be analogous across different polycystine groups. Due to the physical resemblance of two Procyttarium

primordialis? (Collodaria) specimens with the putative juvenile Nassellaria and Spumellaria specimens we observed (S5 Fig), it is possible that this morphotype is common to all polycystine orders during an early juvenile life stage, and not only in the Collodaria. These specimens’ ASV signatures placed them across all three polycystine orders; however, the speci- mens were morphologically indistinguishable. They shared a light, semi-transparent central spherical feature surrounded by algal symbiont cells encased in a gelatinous cytoplasm that could appear clear or slightly striated or cloudy. Although each cell contained a collection of symbionts, there was no siliceous skeletal development in any of them. To our knowl- edge, no such morphotype has previously been described in the literature, and we hypothesize that it could be an interme- diate stage between swarmer cell and skeletal formation during which the specimen acquires photosymbionts. This finding is consistent with other observations of polycystine ontogeny that document skeletal growth (e.g., [5]), but may represent an even younger life stage prior to initial skeletal formation. If there were distinguishing morphological features among these putative juvenile specimens, they were not obvious under our shipboard dissecting microscope (S5 Fig). It is also possible that some taxa (at least within the Collodaria) have more than one life stage that can be described as a naked solitary cell with symbionts: once early in ontogeny and again during reproduction. Reproductive swarmer cells have been

reported in Thalassicolla and other solitary collodarian taxa, suggesting that this morphotype could be a reproductive life stage rather than (or in addition to) a juvenile one [12,61]. However, swarmer release has also been reported in colonial Collodaria specimens [62,63], and in fully-skeletonized Nassellaria and Spumellaria [64]. Thus, naked solitary specimens are not the only polycystine morphotype to exhibit evidence of reproductive swarmers. Our findings underscore the need

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 19 / 29 to examine more such specimens via combined microscopy and nucleic acid sequencing to bridge our current knowledge gaps in radiolarian life cycles and phylogenetic relationships. Metabarcoding presents a rich, but enigmatic picture of protistan diversity Metabarcoding surveys report on the diversity of molecular sequences in a sample or study area (e.g., [7]), but we cur- rently lack a method for relating these molecular signatures to classical morphospecies concepts, which have been the basic unit of much biological research and theory developed over the last two centuries. These molecular species are a new frontier in diversity research and cannot simply be translated into existing morphological, ecological, or biological spe- cies concepts [3]. Despite its limitations, marine metabarcoding has already transformed our understanding of protistan life [2]. These data have the power to reveal the behavior, functions, and diversity of microscopic organisms on a stagger- ing scale that has not been approachable using traditional methods. Although morphospecies are still considered the “gold standard” for measuring protistan diversity, they are ultimately artificial constructs and not necessarily reliable for all groups. For example, it is well-known in the field of radiolarian systematics that some groups are over-split while other groups are over-lumped. Early literature documenting radiolarian

species richness (e.g., [65]) included sometimes vague descriptions of newly-described taxa and occasionally lacked any illustration or type material, leading to subsequent confusion of morphospecies categories. Since then, much work has been done to improve radiolarian morphological taxonomic concepts (e.g., [66]) and develop radiolarians into reliable biostratigraphic markers that are frequently employed in the age determination of marine sediments [67]. These marker taxa (chronological species) are typically large (>100 μm) with distinct morphological characters, and their elevated bio- stratigraphic importance can lead to them receiving great attention to detail (and potential for subspecies-level splitting). Other taxonomic groups have been chronically under-documented, in part due to their irrelevance for biostratigraphy, relatively small size (<100 μm), and complicated morphological characters, leaving many presumed species-level taxa in open nomenclature or in broad species categories (such as the Lophophaenidae [55]). This essentially leaves us with two moving targets: molecular and morphological species concepts, which compounds the difficulty of reconciling them. Our goal in this long-running reconciliation does not need to be molding one type of data to fit neatly into the other; rather, we can let them work together. Morphological and molecular data are both valuable, as they each can shed light on different aspects of protistan biodiversity, life cycles, and ecological roles. By integrating them we will deepen our overall

understanding of these topics in ways that cannot be achieved with morphological or molecular data alone. As we have shown here, it is not a simple task to merge these types of data, due to factors like intragenomic and intraspecies genetic variation, which manifest very differently across taxa. Our results demonstrate that individual polycystine radiolarian spec- imens usually contain a multitude of ASVs, and their ASV assemblages are inconsistent among specimens belonging to the same morphospecies, even those from the same population. Similar findings of genetic diversity have been observed in other protistan groups as well, suggesting that this phenomenon is widespread, highly complex, and can severely obfuscate diversity metrics in metabarcoding studies (e.g., [39,40]). By adopting a flexible approach to metabarcoding data interpretation that is informed by known characteristics of the target genomes, such as intragenomic variation, copy number, and ploidy, we can begin to build a stronger foundation for new discoveries. Materials and methods Approach This study was designed to characterize 18S rRNA gene sequence variability within morphospecies and isolated speci- mens of polycystine Radiolaria. To accomplish this, live radiolarian specimens belonging to orders Collodaria, Nassellaria, and Spumellaria were collected at sea and processed for both microscopy and long-read nucleic acid circular consensus sequencing. Sequence data were brought through a standard bioinformatic pipeline and customized taxonomic filtering

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 20 / 29 procedure to identify the exact 18S amplicon sequence variants (ASVs) and their relative read abundances within solitary and colonial specimens belonging to known morphotaxa. ASVs from groups of specimens belonging to the same mor- phospecies were then used to assess intraspecific genetic variability. A phylogenetic tree was constructed from the long- read ASV data as well as reference sequences, and used in conjunction with ASV presences and abundance patterns to infer relationships between morphotaxa. Sample collection Plankton drift net sampling was conducted at and near Station ALOHA (22° 45’N, 158° 00’W) on Hawaiian Ocean Time-series (HOT) cruises 338 and 339 in July–August 2022 and August–September 2022, respectively. Additional net samples were collected on the Simons Collaboration on Ocean Processes and Ecology-PARticles And Growth in the Oce- anic Nutricline (SCOPE-PARAGON II) cruise in August 2022. All cruises took place on the R/V Kilo-Moana. A SEA-GEAR Model 9000 50 cm diameter plankton net with 50μm mesh was deployed for 5-minute intervals to tar- get solitary Nassellaria and Spumellaria. For the larger Collodaria specimens, a 1m diameter net with 1 mm mesh was used. Samples were collected at a depth of ~5m by-hand with a rope, while a double-release opening-closing mechanism enabled additional net sampling at discrete depths (150m and 300m). Immediately upon recovery, samples were examined onboard using a dissecting microscope (Leica MZAPO) with 0.63x

and 1.6x objectives. Individual polycystine radiolarian specimens were photographed, isolated using a Pasteur pipette, and rinsed 3 times in 0.2μm filtered seawater to remove any attached particles, such as prey or other contaminants introduced during net tow sample collection. Specimens were then transferred to 2 ml cryovials with approximately 0.5 ml of 0.2μm filtered seawater and allowed to incubate at room-temperature for ~4 hours to facilitate prey digestion. Cryovials were then frozen at –80˚C onboard and shipped to the University of Southern California on dry ice. In total, 304 isolated specimens were obtained and processed. Sample processing and primer design DNA was extracted from each specimen using the MasterPure Complete DNA and RNA Purification Kit (Biosearch Technologies) following the manufacturer’s instructions for “fluid samples.” The only modifications were: the debris pel- let was retained for taxonomic identification of the specimen’s skeletal remains, and DNA from solitary specimens (i.e., single-cells) was resuspended in 20uL rather than 35uL TE buffer in order to increase DNA concentration for PCR amplifi- cation (Steps 3 and 8 in Part D of the manufacturer’s instructions, respectively). The extracted DNA concentration of each specimen was measured with a Qubit 2.0 fluorometer. Extracted DNA was PCR amplified using three custom primer pairs designed to target ~1450 bp segments of the 18S rRNA gene in either the Nassellaria, Spumellaria, or Collodaria. Targeted primers (Table 2) were developed by manu-

ally inspecting all curated polycystine reference sequences available in the PR2 database v4.14 as well as a long-read environmental DNA sequence database generated by Jamy et al. [56] Using AliView software [68], radiolarian sequences were visually compared to 80% consensus sequences of all other eukaryotic groups in PR2 (at approximately phylum-class level; PR2 taxonomy level 4) to check the specificity of potential primer sequences. Then, primer pair can- didates were tested in-silico using Silva TestPrime 1.0 and the primer test function in PR2 to confirm that relatively few Table 2. Polycystine radiolarian primers designed for this study. Target group Forward primer (5’-3’) Reverse primer (5’-3’) Nassellaria GTCGGTAGTGTATTGGRCTAC CYGCAGGTTCACCTACGG Spumellaria GACGTKTCATTCAAATTTCTGCC GTTACGACTTCTCCTTCCTC Collodaria GGTTGABCCTGCCAGTAG CYTATTGTAGCCCGTGCGC https://doi.org/10.1371/journal.pone.0326053.t002

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 21 / 29 non-target taxa would be amplified as PCR products. These tests found that with zero mismatches, the primer pair hits were 96% Nassellaria, 76–88% Spumellaria, and 100% Collodaria, indicating that they would theoretically amplify the target taxa with limited contamination from other groups. Extracted DNA of each specimen was amplified using the primer pair designed for the appropriate radiolarian order, based on initial taxonomic assignments during shipboard collection and inspection. Although Collodaria are phylogenetically posi- tioned within Nassellaria, it is common practice among radiolarian workers to consider them a distinct order-level group due to their unique characteristics among polycystines: colonial life habit, all described species have photosymbionts, sometimes cells contain multiple nuclei, and some taxa lack a silicified skeleton [5]. The Collodaria primer set was initially designed to capture family Collosphaeridae and limit amplification of the other collodarian groups (because Collosphaeridae is the only family in which all described species build skeletons that can be morphologically identified). However, in practice, these prim- ers amplified specimens belonging to all collodarian families. Primer sequences were obtained from Integrated DNA Technol- ogies (IDT) with M13 tails and a 5’ block, as recommended by Pacific Biosciences (PacBio) for multiplexed amplicon library

preparation (PacBio publication# 101-921-300, revision 2, April 2022). M13 barcodes were also obtained from IDT to allow for asymmetrically-barcoded samples (up to 384 combinations) compatible with PacBio SMRTbell adapter ligation. Target amplification (PCR-1) was conducted in 25 μL reactions, which included: 12.5 μL KAPA HiFi HotStart Ready- Mix, 2.5 μL forward primer (3μM concentration), 2.5 μL reverse primer (3μM concentration), up to 7.5 μL DNA sample, and 7.5 μL–DNA sample volume of molecular biology grade water. Whenever possible, 10ng of DNA was input into the PCR reaction per individual specimen. This amount was typically only attained for colonial specimens, however, as most solitary specimens yielded very low DNA concentrations. Nassellaria and Spumellaria specimens yielded an average of 2.8ng and 3.6ng total DNA, respectively, whereas Collodaria specimens (most of which were colonial) yielded 250ng DNA on average. In cases where a specimen’s DNA concentration was < 0.75ng/μL, the maximum volume allowable in the reaction (7.5 μL) was used in PCR-1. Even reactions with ~0.5ng total DNA input often yielded successful PCR results, so low amounts of raw DNA did not appear to severely impact downstream processing. The thermocycler program for PCR-1 proceeded as follows: initial denaturation at 95°C for 5 minutes, then 35 cycles of denaturation at 98°C for 30 seconds, annealing for 45 seconds at 54°C (for Nassellaria and Spumellaria) or 64°C (for Collodaria), extension at 72°C for 90 sec- onds, followed by a final extension at 72°C for 2 minutes.

Following the first round of PCR, samples were purified with PacBio paramagnetic beads using standard clean-up pro- tocols. Purified PCR-1 products were Qubit-quantified. Specimens that lacked a detectable amount of DNA at this stage were re-run; if they still did not amplify after a second PCR-1 attempt, they were removed from any further processing and analysis. Out of 304 specimens, 54 failed PCR amplification and 18 were lost to a microcentrifuge tube malfunction, leaving a total of 232 successfully-amplified specimens. These samples underwent a second round of PCR (PCR-2) for barcode attachment. The ingredients in PCR-2 were: 12.5 μL KAPA HiFi HotStart ReadyMix, 1.25 μL forward M13 bar- code (10μM concentration), 1.25 μL reverse M13 barcode (10μM concentration), up to 10 μL cleaned PCR-1 product, and 10 μL–PCR-1 product of molecular biology grade water. PCR-1 products were diluted as needed so that 10ng of DNA was input into PCR-2. The following thermocycler program was used: initial denaturation at 95°C for 5 minutes, then 2 cycles of denaturation at 98°C for 30 seconds, annealing at 60°C for 30 seconds, and extension at 72°C for 90 seconds, followed by 25 cycles of denaturation at 98°C for 30 seconds, annealing at 66°C for 30 seconds, and extension at 72°C for 90 sec- onds, with a final extension at 72°C for 5 minutes. A second PacBio bead clean-up was performed prior to Qubit quantification. PCR-2 products were run on SYBR-safe stained 1% agarose gels to confirm amplification in the size range of the targeted amplicon (~1450 bp) and cleanup of

debris fragments were successful. Sample concentrations were standardized and then pooled. The pool was brought to the University of California, Irvine Genomics Research and Technology Hub, where SMRTbell library preparation was performed by staff using PacBio’s SMRTbell 3.0 prep kit. The finished library was sequenced on a single PacBio Sequel II SMRT Cell 8M.

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 22 / 29 The retained skeletons post-DNA-extraction were gently rinsed in deionized water before being placed onto individual glass coverslips. Once moisture had fully evaporated, coverslips were mounted onto glass microscope slides with Canada Balsam in xylene (Refractive Index 1.52). Specimens were observed under transmitted light (Olympus BX51 compound microscope at 100x, 200x, and 400x) so that skeletal remains could be classified to the most precise taxonomic level possible. All taxonomic classifications were performed by ST based on formal descriptions and illustrations in several published literature sources, including: [11,24,38,49,55,57,60,62,65,67]. The curated online taxonomic resource Mikrotax (www.mikrotax.org [69]) was also consulted to confirm species identifications. If specimens lacked a skeleton (i.e., naked Collodaria or juveniles) or if a skeleton could not be identified due to breakage, taxonomic classification was made to the lowest level possible based on initial shipboard photos. The Collodaria specimens in this study were morphologically iden- tified as Acrosphaera murrayanna (6 specimens), Collosphaera huxleyi (51), Collozoum sp. (6), Disolenia zanguebarica (2), Siphonosphera socialis (6), Procyttarium primordialis? (3), Rhaphidozoum sp. (3), Sphaerozoum sp. (8), or Thalassi- colla nucleata (10). Nassellaria specimens were identified as acanthodesmid sp. (1 specimen), Acanthodesmia vinculata

(2), Eucyrtidium hexagonatum (2), Lophophaena hispida (1), Peromelissa thoracites (1), Pterocanium praetextum (3), Pterocorys campanula (1), Pterocorys sp. (4), Pterocorys zancleus (19), or Tetracorethra tetracorethra (1), and one puta- tive juvenile specimen. Spumellaria specimens were comprised of Dictyocoryne profunda (6 specimens), Dictyocoryne sp. (4), Dictyocoryne truncatum (3), Didymocyrtis sp. (6), Didymocyrtis tetrathalamus coronatus (3), Didymocyrtis tetrathala- mus tetrathalamus (7), Euchitonia elegans-furcata (5), Pyloniidae sp. (4), or lithellid sp. 3 (sensu Trubovitz et al. [55]) (1), and three putative juvenile specimens. Data analyses In total, 5,247,250 barcoded circular consensus reads were received from the University of California, Irvine Genomics Research and Technology Hub, with a mean of 22,617 reads per specimen. These reads were put through the DADA2 bioinformatic pipeline [70] to remove primers, filter by length and quality, denoise errors, call ASVs, and remove chime- ras. Sequences were initially checked for primer matches and oriented in the forward direction using cutadapt, with the “linked adapter” setting (-g) and the parameters --revcomp --discard-untrimmed -e 0.1, to account for reverse- complemented reads, remove reads lacking both forward and reverse primer sequences, and limit the allowable error rate to 10%. PacBio Circular Consensus Sequence data must be corrected for orientation because reads can be output in either the forward or the reverse direction. DADA2 removePrimers function was then applied to double-check read

orientation using the orient = TRUE parameter. The removePrimers function was also used to confirm that all reads contained primer matches (using the default maximum mismatch threshold of 2), remove any reads that did not contain primer matches, and then trim primer sequences from all passing reads. Next, the DADA2 filterAndTrim function was used to filter reads for quality and length. We used the parameters maxLen = 1600 and minLen = 1300, based on the expected length range of our target amplicons (1400–1500 bp). Following the filtering recommendations for long-read PacBio amplicon data [71], we required that reads contain zero Ns, have a minimum quality score of 3, and a maximum expected error rate of 2 (maxN = 0, minQ = 3, maxEE = 2). Within the DADA2 learnErrors function, we used the param- eters “errorEstimationFunction=PacBioErrfun” and “BAND_SIZE=32,” as advised for PacBio long-read data [71]. To best detect rare sequence variants that may be shared by multiple specimens, the parameter pool = TRUE was used in executing the dada function. Finally, the removeBimeraDenovo function in DADA2 was used to identify and remove chimeric reads. These steps were designed to separate spurious reads due to technical errors from authentic biolog- ical sequence variants [70]. Although we implemented the recommended parameters for quality filtering and calling ASVs with PacBio long-read circular consensus sequence data [71], there is no set of parameters that is guaranteed

to remove all PCR and sequencing errors while retaining all authentic biological sequence variants. Thus, it is possible that our stringent filtering methods removed some authentic sequence variants or failed to remove some erroneous reads.

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 23 / 29 Table 3 tracks the mean percent of reads that passed through each stage of the DADA2 pipeline for each order. It is presumed that the relatively low passing rates at the quality and length filtering step is primarily due to length rather than quality. The sequencing facility reported that 100% of our barcoded reads had Phred quality scores ≥Q20. Thus, PCR artefacts or off-target amplification are the most likely explanations for our passing rates. The mean read length of the raw demultiplexed data was 1252 bp; however, based on known genomic characteristics of the target taxa, we expected on-target reads to be approximately 1400–1500 bp long. We thus required that reads passing our length filter be between 1300 and 1600 bp long, which alone would have removed approximately half the raw reads for being too short. These rel- atively short reads may have originated from artefacts generated during PCR, or from contaminating taxa that happened to have shorter regions of the 18S gene and were unexpectedly amplified by our primers. ASV data were then imported into QIIME2 [72] for taxonomic assignment and filtering. In QIIME2, naïve bayes classi- fiers were trained on the PR2 database with the custom primer pairs (Table 2). After taxonomic classifications were made (using minimum confidence = 0.3), any sequences that did not belong to polycystine Radiolaria or had < 3 total reads were filtered out. In an additional effort to remove potential sources of environmental contamination from each specimen, any

ASV that was not assigned to the same family as the morphospecies was excluded from analyses. An exception to this protocol was made for naked Collodaria and putative juvenile specimens, which lacked the skeletons needed for confi- dent morphospecies identifications. For these specimens, ASVs with any polycystine family assignment were permitted. Finally, a per-specimen read abundance threshold was applied, which required each ASV to have at least 3 reads in a given specimen before it was counted as present in that specimen. After these vetting procedures, any specimens lacking a minimum of 5 taxonomically-consistent ASV reads were removed from all analyses. A maximum-likelihood phylogenetic tree (constructed with RAxML-NG [73]) was used to detect patterns of suspi- cious topology that may indicate hidden chimeric sequences. Input data included 258 polycystine radiolarian reference sequences, 18 Acantharia outgroup sequences downloaded from the PR2 database (S1 Table), and the most abundant sequence from each of the specimens in this study (available in S1 Data). Sequences were aligned using MAFFT [74] with linsi parameters and then trimmed with a 5% gap threshold using trimAL [75]. The tree utilized the GTR + G evolution- ary model with 100 bootstrap replicates and 1853 aligned positions. After visual inspection, ASVs that formed long- branching clades without any reference sequences and which had relatively low read numbers (i.e., < 50) were manually checked for chimeras by querying NCBI–BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) with the first ~300 bases and last

~300 bases of each sequence separately. This test led to the discovery of 14 chimeric ASVs that had not been removed by the DADA2 algorithm and were removed manually. All these chimeric sequences had been taxonomically assigned to family Pterocorythidae (Nassellaria) in PR2, although NCBI–BLAST identified their sequence fragments as belonging to various groups, including Rhizaria, stramenopiles, and alveolates. Following all quality checking procedures, 173 out of 232 individual specimens passed through all process- ing steps (75%), comprising 1,224 unique ASVs among them. This included 836 ASVs assigned to Collodaria (95 specimens), 187 ASVs assigned to Nassellaria (36 specimens), and 201 ASVs assigned to Spumellaria (42 specimens). This made up the dataset used for all analyses in this study (S1 Data). There were 59 spec- imens that successfully amplified during PCR but did not pass bioinformatic processing and filtering steps, Table 3. Mean percent of total input CCS reads that passed through each step of the DADA2 pipeline per polycystine radiolarian order. Order Primers trimmed (% input) Quality & length filtered (% input) Denoised (% input) Non-chimeric (% input) Nassellaria 92 37 33 30 Spumellaria 89 35 30 24 Collodaria 90 59 56 53 https://doi.org/10.1371/journal.pone.0326053.t003

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 24 / 29 due to their lack of sufficient reads (<5 reads) belonging to taxonomically-appropriate ASVs. It is interpreted that any apparent amplification was due to off-target environmental DNA that may have come into contact with the specimen prior to or during sample collection, and the target DNA coincidentally failed to properly extract or amplify from the specimen. A similarly-designed study [28] found that environmental contamination can contribute exogenous sequence data when amplifying isolated radiolarian cells, even after cleaning and prey digestion. Thus, while we made our best efforts to remove contamination from every specimen in this study via stringent sample preparation and bioinformatic filtering by both read abundance and ASV taxonomic assign- ment (described above), we cannot absolutely guarantee that every ASV we document here originated from within the target specimen. To compare the genetic richness information provided by our long-read ASVs with commonly used V4 hypervariable region ASVs, we extracted V4 sequences from each full-length ASV. V4 regions were identified by aligning all sequences with MAFFT [74] using lingsi parameters, then searching for matches to the general eukaryotic primers TAR-EukF1 (5’-CCAGCASCYGCGGTAATTCC-3’)/ TAR-EukR3 (5’-ACTTTCGTTCTTGATYRA-3’) [76] using AliView software [68]. ASVs that had identical V4 regions were merged for the purpose of building V4-only occurrence charts (S1 Data; sheets

4–6) and exploring V4-only intragenomic variation (Fig 1, e–h) and intra- and inter-species variation (Fig 7b). After merg- ing identical V4 sequences, there was a total of 604 unique V4 sequences in the dataset; 417 assigned to Collodaria, 70 to Nassellaria, and 117 to Spumellaria. We computed p-distances within and between specimens and morphospecies to provide quantitative metrics of intrag- enomic and intraspecies genetic variation. P-distance is the proportion of sites that are different between two aligned sequences being compared. We used MEGA (Molecular Evolutionary Genetic Analysis) 12 software [50] to calculate the mean p-distances within and between groups of ASVs belonging to individual specimens, and comprising each morphos- pecies. The MEGA analysis option “Compute within group mean distance” was used to obtain average intragenomic and intraspecies p-distances within each polycystine order, and the option “Compute between group mean distance” provided average genetic distances between specimens and morphospecies (Fig 7; S2 Data). Distance calculations were per- formed separately on full-length ASVs and V4-region ASVs for comparison. Only specimens that were morphologically- identified to the species level were included in p-distance calculations, to avoid mixing unequal taxonomic levels in the same analysis. ASV richness, presence, and abundance patterns within specimens, morphospecies, and orders were assessed and visualized in R (v4.3.1, “Beagle Scouts”) [77] using the dplyr [78] and ggplot2 [79] packages. ASVs for each of the 173

specimens were ranked based on decreasing relative abundance per specimen in base R. The most abundant ASV in each specimen was used to construct a maximum likelihood phylogenetic tree in RAxML-NG using the same parameters as the initial tree described above, and was visualized in FigTree v1.4.4 [80] (Fig 8; S4 Fig). This phylogenetic tree was used to infer relationships among specimens collected for this study with previously-published radiolarian sequences as a framework. A flowchart summarizing all bioinformatic processing and analytical steps is given in S6 Fig. Supporting information S1 Fig. Plates of all 173 isolated specimens used in analyses: living and skeletal. Photos taken of live specimens in darkfield and brightfield with the shipboard dissecting scope have been cropped and adjusted for white balance and con- trast when necessary. Photos taken of skeletal remains with the compound microscope are unedited. Specimens collected on the SCOPE-PARAGON II cruise were not photographed live, so only photos of skeletal remains are included. Other specimens that only have one photo either could not be photographed on board the ship due to high rocking motion, or they broke beyond recognition during/after DNA extraction and could only be identified from their live photos. (PDF)

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 25 / 29 S2 Fig. Heatmap illustrating ASV presence and abundance in all 51 Collosphaera huxleyi specimens analyzed in this study. Specimen names are shown on the y-axis, corresponding to specimen names in S1 Data and S2 Fig. The x-axis denotes individual ASVs (represented by tick marks) present in each specimen of the morphospecies. Relative read abundances of ASVs per specimen are indicated by the shade of blue (light = low abundance, dark = high abundance); ASVs not present in a particular specimen are indicated as white background. Representative micrographs of C. huxleyi are shown below the heatmap. (PDF) S3 Fig. Distribution of relative read abundances among the top 10 ranking V4-ASVs per specimen. Dots represent relative V4-ASV abundances (percent of total reads yielded by each specimen) at each designated rank (1st–10th) for all 173 specimens. Boxplots denote the median, interquartile range, maximum, and minimum values (outliers excluded) of V4-ASV relative abundances for each order (Collodaria = pink; Nassellaria = brown; Spumellaria = green). Data underlying this figure is in S1 Data. (PDF) S4 Fig. Expanded phylogenetic tree showing the dominant ASVs of all 173 specimens. Tip labels indicate the morphological taxonomy of specimens with a given dominant ASV, as well as the number of specimens that shared the same dominant ASV. ASV identifiers correspond with those in S1 Data. Polycystine radiolarian reference sequences

downloaded from PR2 are denoted with black text. Green text indicates new ASVs obtained from specimens in this study. Parameters used to generate this tree are described in “Materials and methods.” Accession numbers for the polycystine reference sequences and acantharian outgroup sequences can be found in S1 Table. (PDF) S5 Fig. Putative juvenile specimen morphologies across all polycystine Radiolaria orders, photographed live under darkfield and brightfield microscopy (shipboard dissecting scope; Leica MZAPO). Each box displays spec- imens that shared dominant ASVs and are presumably intraspecific, and/or have a dominant ASV that closely matches a reference sequence in PR2 or GenBank. Background colors indicate the polycystine order determined based on ASV taxonomic assignments (pink = Collodaria; brown = Nassellaria; green = Spumellaria). These solitary specimens lacked any siliceous skeletonization needed for morphological taxonomic classification, but molecular sequencing revealed taxonomic identities in all three orders. This morphotype is therefore hypothesized to be a previously-undescribed juvenile life stage that has similar characteristics in Collodaria, Nassellaria, and Spumellaria. (PDF) S6 Fig. Bioinformatic flowchart, summarizing each step of the data processing and analysis workflow used in this study. The purple box denotes the type of input data, blue boxes show intermediate steps in the workflow, and green boxes indicate the main results discussed in this study. (PDF)

S1 Data. Occurrence chart of all long-read and V4 region ASVs per specimen and their read abundances. There are individual sheets for Collodaria, Nassellaria, and Spumellaria specimens, titled either “full-length ASVs” or “V4-only” respectively. The first column of each sheet lists the ASV identifiers (as discussed in the text and displayed in figures). The second column gives the full ASV sequence. The third column, “Taxon,” provides the taxonomic assignment of each ASV based on PR2 reference taxonomy, and the “Confidence” column lists the confidence level of each molecular taxonomic assignment on a scale from 0–1. Subsequent columns give the specimen morphologi- cal taxonomic assignments with numeric identifiers (as referenced throughout in the manuscript text and in figures), and the absolute number of reads of each ASV produced by individual specimens. All ASVs that matched the family

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 26 / 29 assignment of each morpho-specimen are included, even those with <3 reads per specimen. The sheets containing extracted V4-regions of each ASV indicate which of the original long-read sequences had unique V4 regions ver- sus identical V4 matches with other sequences (causing their read counts to be merged in the occurrence chart). Each merged V4 sequence is labelled “merged” in the ASV_ID column; all other V4 ASVs retain the same IDs as their long-read counterparts. For the analyses in this study, ASVs with <3 reads in each specimen were removed to account for potential contamination (see “Materials and methods”), but the data in raw form is provided here to enable maximum flexibility in future use. (XLSX) S2 Data. P-distances quantifying genetic variation within and between specimens and morphospecies. The first sheet lists intra-specimen mean p-distances (number of base differences per site averaged across all sequence pairs comprising each specimen). P-distances between the groups of ASVs comprising each morphospecies are pro- vided in the matrices in second sheet of the dataset; calculations within each taxonomic order were computed sepa- rately. Likewise, intra-morphospecies and inter-morphospecies p-distances are given in the third and fourth sheets, respectively. When a specimen or morphospecies only yielded a single ASV, p-distance could not be calculated (“N/A” cells in the spreadsheets). Only specimens morphologically-classified to the species level were included in p-distance analyses.

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 27 / 29 References 1. Santoferrara L, Burki F, Filker S, Logares R, Dunthorn M, McManus GB. Perspectives from ten years of protist studies by high-throughput metabar- coding. J Eukaryot Microbiol. 2020;67(5):612–22. https://doi.org/10.1111/jeu.12813 PMID: 32498124 2. Burki F, Sandin MM, Jamy M. Diversity and ecology of protists revealed by metabarcoding. Curr Biol. 2021;31(19):R1267–80. https://doi. org/10.1016/j.cub.2021.07.066 PMID: 34637739 3. Caron DA, Hu SK. Are We Overestimating Protistan Diversity in Nature?. Trends Microbiol. 2019;27(3):197–205. https://doi.org/10.1016/j. tim.2018.10.009 PMID: 30455081 4. Guillou L, Bachar D, Audic S, Bass D, Berney C, Bittner L, et al. The Protist Ribosomal Reference database (PR2): a catalog of unicellular eukary- ote small sub-unit rRNA sequences with curated taxonomy. Nucleic Acids Res. 2013;41(Database issue):D597-604. https://doi.org/10.1093/nar/ gks1160 PMID: 23193267 5. Suzuki N, Not F. Biology and Ecology of Radiolaria. In: Ohtsuka S, Suzaki T, Horiguchi T, Suzuki N, Not F, editors. Marine Protists: Diversity and Dynamics. Tokyo: Springer; 2015:179–222. 6. Lazarus D. World List of Polycystina (Radiolaria) (version 2024-09-01). In: Bánki O, et al. Catalogue of Life (Version 2024-09-25). Catalogue of Life, Amsterdam, Netherlands. Available from: https://doi.org/10.48580/dfrdl-3c9 7. de Vargas C, Audic S, Henry N, Decelle J, Mahé F, Logares R, et al. Ocean plankton. Eukaryotic plankton diversity in the sunlit ocean. Science.

2015;348(6237):1261605. https://doi.org/10.1126/science.1261605 PMID: 25999516 8. Boenigk J, Ereshefsky M, Hoef-Emden K, Mallet J, Bass D. Concepts in protistology: species definitions and boundaries. Eur J Protistol. 2012;48(2):96–102. https://doi.org/10.1016/j.ejop.2011.11.004 PMID: 22209540 9. Santoferrara LF, Grattepanche J-D, Katz LA, McManus GB. Patterns and processes in microbial biogeography: do molecules and morphologies give the same answers?. ISME J. 2016;10(7):1779–90. https://doi.org/10.1038/ismej.2015.224 PMID: 26849313 10. Forster D, Lentendu G, Filker S, Dubois E, Wilding TA, Stoeck T. Improving eDNA-based protist diversity assessments using networks of amplicon sequence variants. Environ Microbiol. 2019;21(11):4109–24. https://doi.org/10.1111/1462-2920.14764 PMID: 31361938 11. Boltovskoy D, Correa N. Planktonic equatorial diversity troughs: fact or artifact? Latitudinal diversity gradients in Radiolaria. Ecology. 2017;98(1):112–24. https://doi.org/10.1002/ecy.1623 PMID: 27935028 12. Anderson OR. Radiolaria. New York: Springer-Verlag. 1983. 13. Michaels AF, Caron DA, Swanberg NR, Howse FA, Michaels CM. Planktonic sarcodines (Acantharia, Radiolaria, Foraminifera) in surface waters near Bermuda: abundance, biomass and vertical flux. J Plankton Res. 1995;17(1):131–63. https://doi.org/10.1093/plankt/17.1.131 14. Lampitt RS, Salter I, Johns D. Radiolaria: Major exporters of organic carbon to the deep ocean. Global Biogeochemical Cycles. 2009;23(1). https:// doi.org/10.1029/2008gb003221

15. Guidi L, Chaffron S, Bittner L, Eveillard D, Larhlimi A, Roux S, et al. Plankton networks driving carbon export in the oligotrophic ocean. Nature. 2016;532(7600):465–70. https://doi.org/10.1038/nature16942 PMID: 26863193 16. Gutierrez-Rodriguez A, Stukel MR, Lopes Dos Santos A, Biard T, Scharek R, Vaulot D, et al. High contribution of Rhizaria (Radiolaria) to vertical export in the California Current Ecosystem revealed by DNA metabarcoding. ISME J. 2019;13(4):964–76. https://doi.org/10.1038/s41396-018- 0322-7 PMID: 30538274 17. Lundgreen RBC, Jaspers C, Traving SJ, Ayala DJ, Lombard F, Grossart H-P, et al. Eukaryotic and cyanobacterial communities associated with marine snow particles in the oligotrophic Sargasso Sea. Sci Rep. 2019;9(1):8891. https://doi.org/10.1038/s41598-019-45146-7 PMID: 31222051 18. Preston CM, Durkin CA, Yamahara KM. DNA metabarcoding reveals organisms contributing to particulate matter flux to abyssal depths in the North East Pacific ocean. Deep Sea Research Part II: Topical Studies in Oceanography. 2020;173:104708. https://doi.org/10.1016/j.dsr2.2019.104708 19. Laget M, Drago L, Panaïotis T, Kiko R, Stemmann L, Rogge A, et al. Global census of the significance of giant mesopelagic protists to the marine carbon and silicon cycles. Nat Commun. 2024;15(1):3341. https://doi.org/10.1038/s41467-024-47651-4 PMID: 38684684 20. Llopis Monferrer N, Boltovskoy D, Tréguer P, Sandin MM, Not F, Leynaert A. Estimating Biogenic Silica Production of Rhizaria in the Global Ocean. Global Biogeochemical Cycles. 2020;34(3). https://doi.org/10.1029/2019gb006286

21. Amaral Zettler LA, Anderson OR, Caron DA. Towards a molecular phylogeny of colonial spumellarian radiolaria. Marine Micropaleontology. 1999;36(2–3):67–79. https://doi.org/10.1016/s0377-8398(98)00028-0 22. Ishitani Y, Ujiié Y, de Vargas C, Not F, Takahashi K. Phylogenetic relationships and evolutionary patterns of the order Collodaria (Radiolaria). PLoS One. 2012;7(5):e35775. https://doi.org/10.1371/journal.pone.0035775 PMID: 22567112 23. Decelle J, Romac S, Sasaki E, Not F, Mahé F. Intracellular diversity of the V4 and V9 regions of the 18S rRNA in marine protists (radiolarians) assessed by high-throughput sequencing. PLoS One. 2014;9(8):e104297. https://doi.org/10.1371/journal.pone.0104297 PMID: 25090095 24. Biard T, Pillet L, Decelle J, Poirier C, Suzuki N, Not F. Towards an Integrative Morpho-molecular Classification of the Collodaria (Polycystinea, Radiolaria). Protist. 2015;166(3):374–88. https://doi.org/10.1016/j.protis.2015.05.002 PMID: 26092634 25. Sandin MM, Pillet L, Biard T, Poirier C, Bigeard E, Romac S, et al. Time Calibrated Morpho-molecular Classification of Nassellaria (Radiolaria). Protist. 2019;170(2):187–208. https://doi.org/10.1016/j.protis.2019.02.002 PMID: 31055251

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 28 / 29 26. Sandin MM, Biard T, Romac S, O’Dogherty L, Suzuki N, Not F. A Morpho-molecular Perspective on the Diversity and Evolution of Spumellaria (Radiolaria). Protist. 2021;172(3):125806. https://doi.org/10.1016/j.protis.2021.125806 PMID: 34174720 27. Krabberød AK, Bråte J, Dolven JK, Ose RF, Klaveness D, Kristensen T, et al. Radiolaria divided into Polycystina and Spasmaria in combined 18S and 28S rDNA phylogeny. PLoS One. 2011;6(8):e23526. https://doi.org/10.1371/journal.pone.0023526 PMID: 21853146 28. Sandin MM, Romac S, Not F. Intra-genomic rRNA gene variability of Nassellaria and Spumellaria (Rhizaria, Radiolaria) assessed by Sanger, Min- ION and Illumina sequencing. Environ Microbiol. 2022;24(7):2979–93. https://doi.org/10.1111/1462-2920.16081 PMID: 35621046 29. Zettler LA, Anderson OR, Caron DA. Insights on the diversity within a “species” of Thalassicolla (Spumellarida) based on 16S-like ribosomal RNA gene sequencing. J Eukaryot Microbiol. 1998;45(5):488–96. https://doi.org/10.1111/j.1550-7408.1998.tb05105.x PMID: 9783450 30. Ishitani Y, Ujiié Y, Takishita K. Uncovering sibling species in Radiolaria: evidence for ecological partitioning in a marine planktonic protist. Mol Phy- logenet Evol. 2014;78:215–22. https://doi.org/10.1016/j.ympev.2014.05.021 PMID: 24862224 31. Suzuki N. Distribution patterns of the radiolarian nuclei and symbionts using DAPI-fluorescence. Bulletin of the National Museum of Nature and Science Series B. 2009;35(4):169–82.

32. Sultanov D, Hochwagen A. Varying strength of selection contributes to the intragenomic diversity of rRNA genes. Nat Commun. 2022;13(1):7245. https://doi.org/10.1038/s41467-022-34989-w PMID: 36434003 33. Hollande A, Cachon-Enjumet M. La polyploïdie du noyau végétatif des Radiolaires. Comptes rendus hebdomadaires des seances de l’Academie des sciences. 1959;248(18):2641–3. 34. Parfrey LW, Lahr DJG, Katz LA. The dynamic nature of eukaryotic genomes. Mol Biol Evol. 2008;25(4):787–94. https://doi.org/10.1093/molbev/ msn032 PMID: 18258610 35. Suzuki N, Aita Y. Radiolaria: achievements and unresolved issues: taxonomy and cytology. Plankton Benthos Res. 2011;6(2):69–91. https://doi. org/10.3800/pbr.6.69 36. Swart EC, Bracht JR, Magrini V, Minx P, Chen X, Zhou Y, et al. The Oxytricha trifallax macronuclear genome: a complex eukaryotic genome with 16,000 tiny chromosomes. PLoS Biol. 2013;11(1):e1001473. https://doi.org/10.1371/journal.pbio.1001473 PMID: 23382650 37. Parks MB, Nakov T, Ruck EC, Wickett NJ, Alverson AJ. Phylogenomics reveals an extensive history of genome duplication in diatoms (Bacillario- phyta). Am J Bot. 2018;105(3):330–47. https://doi.org/10.1002/ajb2.1056 PMID: 29665021 38. Biard T, Bigeard E, Audic S, Poulain J, Gutierrez-Rodriguez A, Pesant S, et al. Biogeography and diversity of Collodaria (Radiolaria) in the global ocean. ISME J. 2017;11(6):1331–44. https://doi.org/10.1038/ismej.2017.12 PMID: 28338675

39. Medinger R, Nolte V, Pandey RV, Jost S, Ottenwälder B, Schlötterer C, et al. Diversity in a hidden world: potential and limitation of next-generation sequencing for surveys of molecular diversity of eukaryotic microorganisms. Mol Ecol. 2010;19 Suppl 1(Suppl 1):32–40. https://doi.org/10.1111/ j.1365-294X.2009.04478.x PMID: 20331768 40. Gong W, Marchetti A. Estimation of 18S Gene Copy Number in Marine Eukaryotic Plankton Using a Next-Generation Sequencing Approach. Front Mar Sci. 2019;6. https://doi.org/10.3389/fmars.2019.00219 41. Ruvindy R, Barua A, Bolch CJS, Sarowar C, Savela H, Murray SA. Genomic copy number variability at the genus, species and population levels impacts in situ ecological analyses of dinoflagellates and harmful algal blooms. ISME Commun. 2023;3(1):70. https://doi.org/10.1038/s43705-023- 00274-0 PMID: 37422553 42. Gong J, Dong J, Liu X, Massana R. Extremely high copy numbers and polymorphisms of the rDNA operon estimated from single cell analysis of oligotrich and peritrich ciliates. Protist. 2013;164(3):369–79. https://doi.org/10.1016/j.protis.2012.11.006 PMID: 23352655 43. Boltovskoy D, Kling SA, Takahashi K, Bjørklund K. World atlas of distribution of recent polycystina (Radiolaria). Palaeontologia Electronica. 2010;13(3):1–230. 44. Greco M, Morard R, Darling K, Kucera M. Macroevolutionary patterns in intragenomic rDNA variability among planktonic foraminifera. PeerJ. 2023;11:e15255. https://doi.org/10.7717/peerj.15255 PMID: 37123000

45. Weber AA-T, Pawlowski J. Wide occurrence of SSU rDNA intragenomic polymorphism in foraminifera and its implications for molecular species identification. Protist. 2014;165(5):645–61. https://doi.org/10.1016/j.protis.2014.07.006 PMID: 25150612 46. Liu K, Huang X, Ding X, Chen N. The high molecular diversity in Noctiluca scintillans is dominated by intra-genomic variations revealed by single cell high-throughput sequencing of 18S rDNA V4. Harmful Algae. 2024;132:102568. https://doi.org/10.1016/j.hal.2024.102568 PMID: 38331542 47. Wang H, Liu K, He Z, Chen Y, Hu Z, Chen W, et al. Extensive intragenomic variations of the 18S rDNA V4 region in the toxigenic diatom species Pseudo-nitzschia multistriata revealed through high-throughput sequencing. Mar Pollut Bull. 2024;201:116198. https://doi.org/10.1016/j.marpol- bul.2024.116198 PMID: 38428045 48. Flegontova O, Lukeš J, Horák A. Intragenomic diversity of the V9 hypervariable domain in eukaryotes has little effect on metabarcoding. iScience. 2023;26(8):107291. https://doi.org/10.1016/j.isci.2023.107291 PMID: 37554448 49. Strelkov AA, Reshetnyak VV. Colonial spumellarian radiolarians of the world ocean. Issledovaniya Fauny Morei. Akademiya Nauk SSSR, Zoolog- icheskii Institut. 1971. 295–418. 50. Kumar S, Stecher G, Suleski M, Sanderford M, Sharma S, Tamura K. MEGA12: Molecular Evolutionary Genetic Analysis Version 12 for Adaptive and Green Computing. Mol Biol Evol. 2024;41(12):msae263. https://doi.org/10.1093/molbev/msae263 PMID: 39708372

51. Morard R, Darling KF, Weiner AKM, Hassenrück C, Vanni C, Cordier T, et al. The global genetic diversity of planktonic foraminifera reveals the structure of cryptic speciation in plankton. Biol Rev Camb Philos Soc. 2024;99(4):1218–41. https://doi.org/10.1111/brv.13065 PMID: 38351434

PLOS One | https://doi.org/10.1371/journal.pone.0326053 June 20, 2025 29 / 29 52. Jensen LØ, Mousing EA, Richardson K. Using species distribution modelling to predict future distributions of phytoplankton: Case study using species important for the biological pump. Marine Ecology. 2017;38(3):e12427. 53. Ezard THG, Aze T, Pearson PN, Purvis A. Interplay between changing climate and species’ ecology drives macroevolutionary dynamics. Science. 2011;332(6027):349–51. https://doi.org/10.1126/science.1203060 PMID: 21493859 54. Lazarus D, Barron J, Renaudie J, Diver P, Türke A. Cenozoic planktonic marine diatom diversity and correlation to climate change. PLoS One. 2014;9(1):e84857. https://doi.org/10.1371/journal.pone.0084857 PMID: 24465441 55. Trubovitz S, Lazarus D, Renaudie J, Noble PJ. Marine plankton show threshold extinction response to Neogene climate change. Nat Commun. 2020;11(1):5069. https://doi.org/10.1038/s41467-020-18879-7 PMID: 33093493 56. Jamy M, Foster R, Barbera P, Czech L, Kozlov A, Stamatakis A, et al. Long-read metabarcoding of the eukaryotic rDNA operon to phylogenetically and taxonomically resolve environmental diversity. Mol Ecol Resour. 2020;20(2):429–43. https://doi.org/10.1111/1755-0998.13117 PMID: 31705734 57. Brandt K. Beiträge zur Kenntnis der Colliden. Archiv für Protistenkunde. 1902;1:59–88. 58. Hollande A, Enjumet M. Contribution à l’étude biologique des Sphaerocollides (Radiolaires Collodaires et Radiolaires polycyttaires) et de leur para- sites. Annales des Sciences Naturelles, Zoologie. 1953;15:99–183.