ISSN 0968-0446 •^•^^ Bulletin o The Natural History Museum HISTORY MUSEUM 01 JOMSI6 PRESENTED GENERAL LIBRARY Botany Series THE NATURAL HISTORY MUSEUM VOLUME 26 NUMBER 1 27 JUNE 1996 The Bulletin of The Natural History Museum (formerly: Bulletin of the British Museum (Natural History)), instituted in 1949, is issued in four scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology. The Botany Series is edited in the Museum's Department of Botany Keeper of Botany: Dr S. Blackmore Editor of Bulletin: Ms M.J. Short Papers in the Bulletin are primarily the results of research carried out on the unique and ever- growing collections of the Museum, both by the scientific staff and by specialists from elsewhere who make use of the Museum's resources. Many of the papers are works of reference that will remain indispensable for years to come. All papers submitted for publication are subjected to external peer review for acceptance. A volume contains about 160 pages, made up by two numbers, published in the Spring and Autumn. Subscriptions may be placed for one or more of the series on an annual basis. Individual numbers and back numbers can be purchased and a Bulletin catalogue, by series, is available. Orders and enquiries should be sent to: Intercept Ltd. P.O. Box 7 16 Andover Hampshire SP101YG Telephone: (01264) 334748 Fax: (01 264) 334058 Claims for non-receipt of issues of the Bulletin will be met free of charge if received by the Publisher within 6 months for the UK, and 9 months for the rest of the world. World List abbreviation: Bull. not. Hist. Mus. Lond. (Bot.) © The Natural History Museum, 1996 Botany Series ISSN 0968-0446 Vol. 26, No. 1 , pp. 1-73 The Natural History Museum Cromwell Road London S W7 5BD Issued 27 June 1 996 Typeset by Ann Buchan (Typesetters), Middlesex Printed in Great Britain at The Alden Press, Oxford Bull. not. Hist. Mus. Lond. (Bot.)26(l): 1-73 Issued 27 June 1996 A morphological study of Chaetoceros species (Bacillariophyta) from the plankton of the Pacific Ocean of Mexico DAVID U. HERNANDEZ-BECERRIL ^ Department of Botany, University of Bristol, Woodland Road, Bristol BS8 1 UG Present address: Laboratorio de Invertebrados, Facultadde Ciencias, UN AM, Apdo. postal 70-371, Coyoacdn, Mexico, D.F. 04510 Mexico CONTENTS NATURAL HISTORY MUSEUM 01 JUL 1993 Introduction General Historical background and taxonomic relationships Recent studies with electron microscopy 1 1 2 2 Materials and methods 2 Source of material 2 Methods 3 Terminology 3 Systematic account 4 Discussion 67 Morphology 67 Taxonomic relationships 68 Classification 68 Biogeography 70 References 71 Index .. ..73 SYNOPSIS. An analysis of 65 preserved marine plankton samples from waters of the Pacific Ocean of Mexico was undertaken in order to investigate the morphology of species of the diatom genus Chaetoceros Ehrenb. Forty-six Chaetoceros taxa were recorded from the samples. Of these, 35 were studied using light, scanning and transmission electron microscopy; a further 1 1 species were also recorded as present in the material, but these have been treated elsewhere. Most of the taxa studied are characteristic of warm-water regions. Their morphology is discussed, especially where it differs from that previously described. Most of the taxa are illustrated by light and electron micrographs. A new section, Peruviana Hern. -Bee., within subgenus Chaetoceros (Phaeoceros), is proposed. New observations by electron microscopy are contributed for the following taxa: C. atlanticus var. neapolitanus, C. atlanticus var. skeleton, C. brevis, C. compressus, C. densus, C. denticulatus, C. distorts, C. pendulus, and C. vistulae. Phylogenetic relationships with other genera and aspects of distribution and biogeography are also revised. INTRODUCTION General Chaetoceros species are an important and often abundant component of the marine phytoplankton community; however, the great number of species makes positive identification difficult. The major characters traditionally used for identification at the species level are: the form and relative length of the chain, the frustule shape (girdle and valve views) and size, the structure and size of the girdle and valve mantle, the orientation and form of the setae, and the number of chloroplasts. Evensen & Hasle (1975) found that the structure and size of the rimoportula is another important feature for identifying species. Hargraves (1979) considered that resting spores can also aid identification, although these are limited to species belonging to subgenus Hyalochaete. Many Chaetoceros species from the plankton of temperate and cold waters have been described and studied using electron microscopy (EM). However, tropical and subtropical species have been less studied using such modem techniques. This work is a study of the morphology and taxonomy of some species of the genus Chaetoceros which were encountered in tropical and subtropical areas, in order to present a systematic account of the species from those regions and contribute to the taxonomy of the genus. © The Natural History Museum, 1996 D. U. HERNANDEZ-BECERRIL Historical background and taxonomic relationships The genus Chaetoceros was erected by Ehrenberg (1844), who described the type species, C. dichaeta, as well as three allied genera: Dicladia, Goniothecium, and Syndendrium; the last two were at spore stage and all three are now considered to be synonyms of Chaetoceros (VanLandingham, 1968). Later, Brightwell (1856) provided an emended description of Chaetoceros, including two new species. Schiitt (1895) described a new genus, Peragallia, which is also now considered to be a synonym of Chaetoceros (probably C. peruvianus Brightw.). Gran (1897) divided Chaetoceros into two subgenera: Phaeoceros and Hyalochaete (Hendey, 1964, has correctly substituted Chaetoceros for Phaeoceros, as this subgenus contains the type species). Ostenfeld (1903) divided the subgenera into 16 sections (two in Chaetoceros and 14 in Hyalochaete); further additions were made by Gran (1905). Recent additions to the classification include the proposal of section Coarctata within subgenus Chaetoceros (Hernandez-Becerril, 199 la), section Conspicua within subgenus Hyalochaete (Hernandez- Becerril et al., 1993), and the proposal for a new subgenus, Bacteriastroidea (Heraandez-Becerril, 19930). The present infrageneric classification includes three subgenera and 19 sections (including the one proposed here): four in subgenus Chaetoceros and 1 5 in subgenus Hyalochaete. According to VanLandingham (1968) there are 177 species of Chaetoceros, including one fossil, four described as spore stages, and five from freshwater environments, together with 36 infraspecific taxa: 19 varieties and 17 forms. However, Hargraves (1979) considered that this number was an overestimation and some species should be regarded as forms or varieties of others. Chaetoceros exhibits little variation in its general structure. The subgenera may be distinguished by the following features: I. Chaetoceros (Phaeoceros)-robust species; one pair of setae per valve; setae thick and strong, most with chloroplasts and spines; resting spores known only in one species. II. Hyalochaete - delicate species; one pair of setae per valve; setae thin and fragile, usually lacking chloroplasts; resting spores in most of the species. III. Bacteriastroidea - fairly robust species; three pairs of setae per valve (two pairs reduced), lacking chloroplasts; resting spores unknown. Evensen & Hasle (1975) noted that the rimoportulae in subgenus Hyalochaete are shorter and less prominent than those in subgenus Chaetoceros, and are confined to terminal cells. The genus Chaetoceros has been considered to belong to the family Chaetoceraceae Smith, together with Bacteriastrum Shadbolt and Acanthoceros Honigm. (Simonsen, 1979). However, Round et al. (1990) placed Chaetoceros within the family Chaetocerotaceae Ralfs, with Bacteriastrum and Gonioceros H. Perag. & Perag., although Crawford et al. (1994) later transferred the two Gonioceros species to the genus Attheya West, leaving the family with only two genera: Chaetoceros and Bacteriastrum. Other workers have included only Chaetoceros in Chaetoceraceae (e.g. Lebour, 1930; Hendey, 1964), but it is evident, however, that Chaetoceros and Bacteriastrum are closely related (Fryxell, 1978; Simonsen, 1979). Acanthoceros is now placed within its own family, Acanthocerataceae R.M. Crawford (Round etal., 1990). In Simonsen's (1979) classification, Chaetoceraceae is placed within the suborder Biddulphiineae, order Centrales. In Round et al. (1990) Chaetocerotaceae is placed within the order Chaetocerotales, subclass Chaetocerotophycidae (both taxa proposed by Round & Crawford in Round et al., 1990); the present study follows this latter classification. Recent studies with electron microscopy Studies of Chaetoceros using electron microscopy began with Okuno (1951), when he published microphotographs of C. criophilus Castrac. Helmcke & Krieger (1953) also made TEM observations of Chaetoceros species, as did Hendey et al. (1954). Desikachary & Bahadur (1954) provided a study of the genus, although gave few details. Desikachary (1956) attempted to summarize the electron microscopy studies (mainly TEM) made on diatoms up to that time. However, his list of 1 1 species of Chaetoceros did not include C. densus Cleve, which had been studied previously by Hendey et al. (1954). Okuno (1956) provided additional information on some partially described species. Later, Hendey (1959) prepared another list, this time including 19 species of Chaetoceros studied using TEM and especially with regard to the cell wall. Okuno (1970) continued to provide more information on Chaetoceros, while Blasco (1970) made a detailed study of C. didymus Ehrenb. Other observations combining LM and TEM were made during the study of life cycles (Hargraves, 1972; Stosch et al., 1973) and morphology (Duke et al., 1973). The most important recent contribution to our knowledge of the genus Chaetoceros was made by Evensen & Hasle (1975), who studied 14 species (four in subgenus Chaetoceros (Phaeoceros), 10 in subgenus Hyalochaete) using SEM and TEM. Later, Fryxell (1978) studied three species of Chaetoceraceae, including two Chaetoceros species, with SEM. Hargraves (1979) provided an important contribution on the resting spores of Chaetoceros by combining LM, SEM, and TEM. Fryxell & Medlin (1981) studied three Chaetoceros species using LM and SEM, especially with regard to chain formation. Navarro (1982) listed and figured 25 species and one variety of Chaetoceros using LM, but gave no new data on their cell structure. Takano (1983) described a new species, C. salsugineus Takano, accompanied by electron micrographs. Koch & Rivera (1984) studied nine species of subgenus Chaetoceros (Phaeoceros), two of them not previously studied with EM. Additional observations have been made recently on valve development (Li & Volcani, 1985), morphology (Rogerson et al., 1986; Giuffre & Ragusa, 1988; Hernandez-Becerril, 199 la, b, 19920, 19930; Moreno Ruiz et al., 1993), taxonomy (Rines & Hargraves, 1986; Hernandez-Becerril et al., 1993), and resting spores (Stockwell & Hargraves, 1986). One of the latest contributions to the knowledge of the genus was made by Rines & Hargraves (1988), who gave an account of species encountered in Rhode Island, U.S.A., using LM, but also discussing important taxonomic points within the genus as a whole. Other studies with EM involve the discovery of more than one rimoportula per valve (Rines & Hargraves, 1990; Hernandez-Becerril, 1991a, c; Hernandez-Becerril et al., 1993) and the description of new species (Hernandez-Becerril, 1991c, 19926). MATERIALS AND METHODS Source of material This study was carried out using preserved marine plankton samples, some collected by the author, but the majority supplied STUDY OF CHAETOCEROS SPECIES by the following colleagues: R. Cortes-Altamirano (ICMyL-UNAM), S. Gomez-Aguirre (IB-UNAM), and institutions: CIB and CICIMAR (La Paz, B.C.S., Mexico). Sixty-five samples were obtained from the tropical-subtropical Pacific Ocean of Mexico: 29 from the Gulf of California, 31 from off the coast of Baja California, five from off the Pacific coast of Mexico (see Table 1). Methods All material prepared for light and electron microscopy was rinsed at least five times with distilled water. Permanent slides for light microscopy were made from the rinsed and cleaned material. The method for cleaning diatoms followed that used by Simonsen (1974) and Hasle (1978), which is basically the oxidation of the organic material (using KMnO4) followed by acid treatment (HC1). Identification, preliminary observations, and measurements were made with an Olympus CH light microscope, phase contrast, and an Olympus Bh with attached camera, phase contrast, bright field and differential interference contrast. Rinsed and cleaned material was used for SEM. Drops of the prepared material were put onto coverslips, air-dried, and then coated with gold in a coating unit. Some previously identified specimens were isolated with the aid of a micropipette, and afterwards the above method was followed. Critical point drying was used for a few samples, after rinsing and dehydration. The scanning electron microscope used was a Phillips 501, usually operated at 10-12kv. For transmission electron microscope studies, only cleaned material was used. This was carefully placed with a micropipette onto copper, Formvar-coated grids, air-dried, and then observed directly. The instrument was a JEOL 1200 EX. Terminology The terminology adopted in this study generally follows that proposed by Anonymous (1975) and Ross et al. (1979). Specific terminology for Chaetoceros follows Brunei (1966, 1972) and Rines & Hargraves (1988). However, the following terms used in this paper have been modified from those cited (Hernandez-Becerril, 199 la, c, I992a): - The term chain is used instead of colony, as colony implies a more complex organization with some interaction between its members. - Chains are arbitrarily considered short if the number of cells in them is less than 10, long if 1 1 cells or more. Usually, there is no overlap between these two categories, as species are typically long-chained or short-chained. - The term heteropolar is applied in the sense of Rines & Hargraves (1988), where the two ends of a chain or the two valves of a cell are markedly different from each other (PI. 1 , Figs 3a, b). Thus, to distinguish each end the terms anterior and posterior are used: the anterior end having valves with curved setae, directed towards the chain or cell, and the posterior end with setae directed away from the chain or cell. - Girdle view is regarded as the broad girdle view, unless otherwise stated (PI. 1, Fig. 1). - 'Aperture' is the space between sibling cells in a chain (PI. 1, Fig. 1). This is described as narrow if it is less than 5 um, and as wide if more than 5 um. A few species show a rather wide range of variation in the size of aperture and in these cases the average measurement is given. The term aperture is Table 1 Plankton samples used for this study. Sample no. Location Cruise Source Coast of Baja California CICIMAR-CIB 8508 CIB 01 26° 41' N 112° 39' W 02 25° 53' N 112° 58' W 03 25° 43' N 113° 17' W 04 26° 10' N 113° 09' W 05 26° 00' N 113° 18' W 06 25° 50' N 113° 46' W 07 26° 27' N 113° 19' W 08 26° 18' N 113° 38' W 09 26° 07' N 113° 57' W 10 26° 25' N 113° 50' W 11 - 12 26° 14' N 114° 29' W 13 26° 52' N 1 14° 01' W 14 26° 24' N 114° 20' W 15 26° 32' N 114° 40' W Gulf of California GOLCA8606 CIB 16 27°28'N111°25'W 17 27° 45' N 110° 53' W 18 27°3rNlll°20'W 19 27° 49' N 1 1 1° 55' W 20 27° 59' N 111° 36' W 21 28°13'N111°14'W 22 28°24'N111°52'W 23 28° 38' N 112° 28' W Coast of Baja California CIB-CICIMAR 8605 CIB 24 26° 25' Ml 14° 1 7' W 25 26° 35' N 113° 58' W 26 26°15'N114°36'W 27 26° 29' N 113° 34' W 28 26° 19' N 113° 48' W 29 26° 03' N 113° 39' W 30 26° 13' N 113° 19' W 31 26° 23' N 113° 00' W 32 25° 47' N 112° 38' W 33 25° 37' N 112° 57' W 34 25° 09' N 113° 05' W 35 24° 45' N 112° 25' W 36 24° 01' N 112° 23' W 37 24° 19' N 112° 34' W 38 23° 55' N 1 1 1° 05'W 39 23°27'N111°12'W Gulf of California GOLCA8411 CICIMAR 40-49 - Gulf of California CORTES II (1985) ICML 50 25° 02' N 108° 32' W 51 25° 33' N 110° 59' W 52 26° 51' N 110° 06' W 53 28° 10' N 112° 48' W 54 29° 12' N 112° 31' W 55 29° 48' N 114° 20' W 56 31° 17' N 114° 22' W 57 28°09'N111°41'W 58 26° 59' N 111° 50' W 59 23° 08' N 109° 27' W 60 21° 38' N 106° 31' W Pacific Ocean coast of CORTES II (1985) ICML Mexico 61 21° 28' N 105° 20' W 62 22° 45' N 108° 53' W 63 20° 49' N 105° 41' W 64 15°46'N95°38'W Author Author 65 17° 22' N 102° 51' W Author Author CIB = Centre de Investigaciones Biologicas de Baja California Sur, A.C., La Paz, B.C.S., Mexico. CICIMAR = Centra Interdisciplinary de Ciencias del Mar (IPN), La Paz, B.C.S., Mexico. ICML = R. Cortes-Altamirano, Institute de Ciencias del Mar y Limnologia (UNAM), Estacion Mazatlan, Sin., Mexico. D. U. HERNANDEZ-BECERRIL chosen in preference to 'foramen' (e.g. Rines & Hargraves, 1988), because the foramen represents a structure in the valve, while the aperture is not considered to be a valve structure. - Setae have been named according to their position in the chain and their position in the cell (Brunei, 1966). However, this terminology is only partly followed here, as it is very complex, i) Position in the chain: 'terminal' is used in the sense of Takano (1983) as the setae in the terminal valve, not the terminal cell, in the chain; 'intercalary' refers to setae of the intercalary valves and cells, including the internal valve of the terminal cell; 'special' is used to describe intercalary setae which are thicker and larger than the others, fused, and with bigger spines, ii) Position in the cell: 'isovalvar', the setae in the same valve; 'heterovalvar', the setae in different valves (PI. l,Fig. 1). - The term 'annulus' (Stosch, 1977) is used to name the 'central hyaline field', as used by Evensen & Hasle (1975) and the 'central area' in the sense of Okuno (1956). - The measurements given for Chaetoceros species are: apical axis (typically the width in the broad girdle view, abbreviated a.a.), pervalvar axis (p.a.), and aperture (ap.) (PI. 1 , Figs 1 , 2). SYSTEMATIC ACCOUNT Thirty-five taxa belonging to the genus Chaetoceros are described, the taxonomic arrangement following the traditional classification (e.g. Hustedt, 1930; Cupp, 1943; Hendey, 1964; Rines & Hargraves, 1988). The local distribution (in the material studied) is indicated by the sample numbers (see Table 1). CHAETOCEROS Ehrenb. in Ber. Akad. Wiss. Berlin 1844: 200 ( 1 844). Type species: Chaetoceros dichaeta Ehrenb. I. Subgenus CHAETOCEROS (Phaeoceros Gran) 1 . Section ATLANTICA Ostenf. Centre of valve with prominent, tubular protrusion of the rimoportula. Apertures wide. 1 . Chaetoceros atlanticus var. neapolitanus (Schrod.) Must. Hustedt, 1930: 645, fig. 366; Hendey, 1937: 290; Cupp, 1943: 104, fig. 59-Be, non d; Sournia, 1968: 46. Chaetoceros neapolitanus Schrod. Schroder, 1900: 29, pi. 1, fig. 4; Hendey, 1964: 1 19, pi. 16, fig. 3. Pis 2, 3. LM. Chains straight, short, rather robust. In girdle view, cells rectangular, the pervalvar axis being longer, the corners smooth, slightly curved. Valve face slightly convex, bearing a process in the middle, mantle low, lines of girdle straight; apertures wide, hexagonal. In valve view, cells elliptical, the setae not diverging widely. Terminal and intercalary setae undifferentiated, generally thick, robust, and spine-bearing; intercalary setae arise from the valve corners, curving smoothly at an angle of about 45° from the chain axis, terminal setae arise at a similar angle from the corners. Numerous small chloroplasts present in cell and setae. Measurements: 7-13 um a.a., 17-25 jam p.a., 24-30 umap. (PI. 2, Figs 1,2) EM. The general morphology of this species is relatively simple. The valve face, mantle, and setae bases are irregularly perforated by small poroids, although weak costae are sometimes apparent close to the bases of the setae (PI. 2, Figs 3-6; PI. 3, Fig. 4). The valve edge shows a hyaline rim and in the mantle rows of thickening run parallel to the lines of the girdle insertion (PI 3, Figs 1 , 2). Rimoportulae are present in each valve in the chain, located in the centre of the valve face. Externally, these are simple tubes open to the outside (PI. 3, Figs 1-3). The structure of the setae changes at their very base and also distally. They are circular in cross-section at the base, perforated by poroids as in the valve (PI. 3, Figs 1, 2). They then curve and fuse with sibling setae, beyond which point they become thicker, square in cross-section, and spine-bearing; the spines run in rows along each edge and the areola pattern comprises two striae between two costae (PI. 2, Fig. 7; PI. 3, Figs 5-7). DISTRIBUTION. 29, 42, 63. REMARKS. See following variety. 2. Chaetoceros atlanticus var. skeleton (F. Schutt) Hust. Hustedt, 1930: 643, fig. 365; Cupp, 1943: 104, fig. 59-Bb, c; Sournia, 1968:46. Chaetoceros skeleton F. Schutt Schutt, 1895: 45, pi. 5, fig. 19; Peragallo & Peragallo, 1897-1 908: 482, pi. 134, fig. 9. ? Chaetoceros polygonus F. Schutt Schutt, 1895: 46, pi. 5, fig. 24; Hustedt, 1920: pi. 323, figs 6, 8; Lebour, 1930: 1 14, fig. 80; Hendey, 1964: 120, pi. 14, fig. 1 (non C. polygonus F. Schutt sensu Hust., 1920: pi. 322, figs 5, 6 = C. atlanticus var. neapolitanus (Schrod.) Hust.). Pis 4, 5. LM. Chains straight, short, delicate. In girdle view, cells octagonal, the apical axis being longer. Valve face convex, bearing a short, cylindrical process, mantle very low, lines of girdle straight; apertures wide. In valve view, cells elliptical, the setae only slightly diverging. Setae thick, coarse, bearing minute spines; terminal setae essentially undifferentiated from intercalary setae, the latter diverging and curving more widely, almost parallel to the chain axis. Several small chloroplasts present in cell and setae. Measurements: 10-17 um a.a., 8-1 1 urn p.a., 8-10 um ap. (PI. 4, Figs 1, 2) EM. The valves are not as heavily silicified as those of var. neapolitanus, but the structure is otherwise similar to that described above. The costa pattern becomes more apparent in this variety and radiates from the centre of the valve (PI. 4, Fig. 4). A hyaline rim is also present on the valve edge and thickenings in the mantle run along the lines of the girdle (PI. 4, Fig. 6). Each valve bears a rimoportula, located in the centre (PI. 5, Fig. 3). This resembles that of the preceding variety, but is shorter (PI. 4, Figs 4, 6; PI. 5, Fig. 1). Inside the rimoportula is a simple hole without a well-developed labiate structure (PI. 5, Fig. 2). The structure of the setae is the same as that described for C. atlanticus var. neapolitanus. One remarkable character is the presence of thicker, large poroids at the base of the setae below the point of fusion (PI. 5, Fig. 6). The areola pattern also comprises two striae between two costae and this is still apparent close to the tip, which is sharp (PI. 5, Figs 4, 5, 7). DISTRIBUTION. 42, 45, 48, 52. Pervalvar Axis Valve Face Girdle Aperture 1. Intercalary Setae Terminal Setae Transapical Axis \^. — Plate 1 Fig. 1 : diagram illustrating terminology used to describe a typical chain of Chaetoceros (broad girdle view). Fig. 2: valve view of a species of Chaetoceros. Figs 3a, 3b: examples of a heterovalvar species: C. peruvianus (3a), and a heteropolar species: C. coarctatus (3b). D. U. HERNANDEZ-BECERRIL Plate 2 Chaetoceros atlanticus var. neapolitanus. Figs 1 , 2: typical chains. LM. Fig. 3: complete chain. SEM. Fig. 4: whole cell. TEM. Fig. 5: valve showing setae and rimoportula. TEM. Fig. 6: detail of chain. SEM. Fig. 7: detail of seta. TEM. Figs 1-3, bar = 20 urn; Figs 4-6, bar = 10 urn; Fig. 7, bar = 2 nm. STUDY OF CHAETOCEROS SPECIES • Plate 3 Chaetoceros atlanticus van neapolitanus. Fig. 1 : part of valve with broken seta and rimoportula. TEM. Fig. 2: part of valve showing setae and rimoportula. SEM. Fig. 3: detail of rimoportula. SEM. Fig. 4: another cell showing some thickening in the mantle. SEM. Fig. 5: seta with rows of spines. SEM. Fig. 6: fusion of two sibling setae. TEM. Fig. 7: seta showing poroid areolae. TEM. Figs 1-3, 5, 7, bar = 2 urn; Fig. 4, bar = 10 urn; Fig. 6, bar = 5 urn. D. U. HERNANDEZ-BECERRIL Plate 4 Chaetoceros atlanticus var. skeleton. Figs 1 , 2: part of chain and complete chain, respectively. LM. Fig. 3: part of chain. SEM. Fig. 4: two sibling valves showing rimoportulae. TEM. Fig. 5: detail of chain. SEM. Fig. 6: one valve showing hyaline rim and thickening in mantle. SEM. Figs 1-3, bar = 20 urn; Figs 4-6, bar = 5 urn. REMARKS. Chaetoceros atlanticus var. neapolitanus and var. skeleton are morphologically very similar to the type, as indicated by Evensen & Hasle (1975) and Koch & Rivera (1984), and their taxonomic position is in little doubt. They differ in having a more temperate or subtropical distribution. Contrary to Simonsen's (1974) opinion that distinction 'seems to be senseless' between these two taxa, I consider that their taxonomic separation should be retained. 3. Chaetoceros dichaeta Ehrenb. Ehrenberg, 1844: 200; Hustedt, 1920: pi. 326, fig. 8; Hustedt, 1930: 648, fig. 367; Hendey, 1937: 291, pi. 6, figs 9, 10; Cupp, 1943: 106, fig. 60; Hendey, 1964: 119, pi. 13, fig. 1; Okuno, 1970: pi. 652; Evensen & Hasle, 1975: 157, figs 1-5; Koch & Rivera, 1984: 64, figs 6-1 2. Chaetoceros janischianus Castrac. STUDY OF CHAETOCEROS SPECIES Plate 5 Chaetoceros atlanticus var. skeleton. Fig. 1 : terminal valve. TEM. Fig. 2: inside view of valve showing rimoportula internally. SEM. Fig. 3: single cell. SEM. Fig. 4: detail of seta with poroid areolae. TEM. Fig. 5: tip of seta. TEM. Fig. 6: point of fusion in two sibling setae. TEM. Fig. 7: detail of another seta. TEM. Fig. 1 , bar = 5 \an; Figs 2, 4-7, bar = 2 um; Fig. 3, bar = 10 nm. 10 Castracane, 1886:77. Pis 6, 7. LM. Chains straight, of variable length. In girdle view, cells rectangular, nearly octagonal, corners smooth. Valve face generally flat, with a thin, cylindrical process at the centre, mantle low, lines of girdle straight; apertures very wide. In valve view, cells nearly circular or elliptical. Setae thick, coarse; intercalary setae curve smoothly and fuse further out from the cells than terminal setae; terminal setae curve towards the chain axis. Numerous, small chloroplasts present in cell and setae. Measurements: 6- 11 uma.a., ll-12ump.a., 1 3-22 um ap. (PL 6, Figs 1,2) EM. Short chains were observed (PI. 6, Fig. 3). The valve appears to be lightly silicified and perforated by small poroids thoughout; no costae were observed (PI. 6, Figs 4-6). Hair-like structures are present on the valve edge, some relatively long, others shorter (PI. 7, Fig. 1). Hyaline rims occur on the edge of the valve and in the mantle. The presence of thickening was also noted in this species (PI. 6, Fig. 6). The rimoportula consists of a simple, cylindrical tube, open to the outside, and with no labiate structure inside (PI. 7, Figs 1-3, 5). Setae differ from the preceding varieties of C. atlanticus chiefly in the areola pattern, which has longitudinal rows of big poroids (PL 7, Figs 3, 6, 8), but no costae. Their polygonal shape in cross-section is also different, with five or six sides, and the rounded tips have a small hole (an opening?) (PL 7, Figs 4, 7). DISTRIBUTION. 12,48. REMARKS. Evensen & Hasle (1975) commentated on the presence of weak costae in the valves of this species, a characteristic not found in the material used for this study. The hair-like structures on the valve edge have been found by some authors (e.g. Evensen & Hasle, 1975; Koch & Rivera, 1984). These same authors have also stated that the seta in C. dichaeta is circular in cross-section, which differs from my observations where it has clearly been seen as polygonal. 2. Section BOREALIA Ostenf. Tubular rimoportula usually absent. Apertures narrow. 4. Chaetoceros octagonus Hern. -Bee. Hernandez-Becerril, 19926: 218, figs 1, 3-15. DISTRIBUTION. 7, 12. 5. Chaetoceros densus Cleve Cleve, 1901: 299; Meunier, 1913: 14, pi. 1, figs 14-28; Hustedt, 1920: pi. 325, fig. 9; Hustedt, 1930: 651, fig. 368; Lebour, 1930: 115, fig. 81; Cleve-Euler, 1951: 96, figs 184a-c; Hendey et al., 1954: 28, pi. 1, fig. 1; Hendey, 1964: 120, pi. 17, fig. 1; Drebes, 1974: 64, figs 48a, b; Rines & Hargraves, 1988: 50, figs 100, 101. Chaetoceros borealis var. densa Cleve Cleve, 1897: 20, pi. 1, figs 3, 4. Pis 8, 9. LM. Chains straight, usually long, robust. In girdle view, cells rectangular or nearly eight-sided, with smooth corners and constrictions at girdle zones; apertures narrow. Valve face flat or slightly convex, mantle high, lines of girdle straight. In valve view, cells elliptical, the setae diverging at an angle of about 30° D. U. HERNANDEZ-BECERRIL from the apical axis. Setae thick, coarse, bearing spines; terminal setae differ from intercalary setae only in direction of divergence, intercalary setae arising almost parallel to the apical axis, terminal setae curving to the chain axis. Numerous, small chloroplasts present in cell and setae. Measurements: 38-45 um a.a., 32^8 um p.a., 3^ um ap. (PL 8, Fig. 1) EM. The valves follow the general pattern of structure for the genus: heavily silicified and perforated by poroids, but here a weak costa pattern is observed and the poroids become scarse close to the valve centre (PL 8, Figs 2-4). The annulus is an oblong hyaline field in the valve centre (PL 9, Figs 2-4). The rimoportula is located at the very centre of the valve, being a simple hole inside with a small protrusion outside (PL 8, Fig. 5; PL 9, Figs 1-4). At the very base of the setae, the wall is perforated by a fine mesh of poroids, resembling a certain type of areola (PL 9, Figs 1 , 4). The setae are almost circular close to the base, but become four-sided after a short distance (PL 9, Figs 5. 6). They bear rows of alternate spines on the edges and the areola pattern is of one stria between two costae (PL 8, Figs 6, 7; PL 9, Figs 5, 6). DISTRIBUTION. 16,42,45,48. REMARKS. This species is closely related to C. borealis Bailey, an opinion which may be confirmed by comparing Evensen & Hasle 's (1975) observations of C. borealis with those made in this study on C. densus. However, two major differences between these species are the position of the rimoportula (in C. densus it is central and in C. borealis it is located to one side) and the presence of poroids in the setae bases, characteristic of C. densus but lacking in C. borealis. 6. Chaetoceros tetrastichon Cleve Hernandez-Becerril, 1992a: 367, figs 1-8. DISTRIBUTION. 12. 7. Chaetoceros seychellarus G. Karst. Hernandez-Becerril, 1993a: 121, figs 16-27. DISTRIBUTION. 18, 19, 45, 48, 63. 8. Chaetoceros denticulatus Lander Lauder, 1864: 79, pi. 8, fig. 9; Hustedt, 1920-1921: pi. 324, fig. 7, pi. 337, fig. 7; Allen & Cupp, 1935: 135, fig. 53; Subrahmanyan, 1946: 129, figs 1 88-190; Sournia, 1968: 47, pi. 3, fig. 24; Simonsen, 1974: 31; Desikachary &Prema, 1987: pi. 258, fig. 5. Chaetoceros nanodenticulatus Okamura Okamura, 1907:91. Chaetoceros denticulatus f. angusta Hust. Hustedt, 1920: pi. 324, fig. 5 Chaetoceros denticulatus f. lata Hust. Hustedt, 1920: pi. 324, fig. 6. PL 10. LM. Chains straight, of variable length, robust. In girdle view, cells octagonal, with smooth corners and constrictions at girdle level. Valve face conical, mantle high, lines of girdle straight; apertures wide. In valve view, cells elliptical, the setae diverging at an angle of about 1 5-20° from the apical axis. Setae arise from close to the valve centre, thick and coarse, with a tooth-like projection near the base in contact with that of the sibling setae. STUDY OF CHAETOCEROS SPECIES 11 Plate 6 Chaetoceros dichaeta. Figs 1 , 2: typical chains. LM. Fig. 3: part of chain. SEM. Fig. 4: valve with long, central rimoportula. SEM. Fig. 5: another valve showing rimoportula and setae. TEM. Fig. 6: complete cell. SEM. Figs 1 , 2, bar = 20 nm; Fig. 3, bar = 10 um; Figs 4-6, bar = 5 jim. 12 D. U. HERNANDEZ-BECERRIL Plate 7 Chaetoceros dichaeta. Fig. 1 : valve showing hair-like structures (arrows). SEM. Fig. 2: inside view of valve. Arrow indicates the rimoportula internally. SEM. Fig. 3: detail of rimoportula (arrow) and broken seta. TEM. Fig. 4: base of seta. SEM. Fig. 5: valve with long hair-like structures (arrow). TEM. Fig. 6: poroid areolae in seta. TEM. Fig. 7: tip of seta. SEM. Fig. 8: tip of seta. TEM. Figs 1, 5, bar = 5 urn; Figs 2-4, 6-8, bar = 2 urn. STUDY OF CHAETOCEROS SPECIES 13 Plate 8 Chaetoceros densus. Fig. 1 : typical chain. LM. Fig. 2: part of chain. SEM. Fig. 3: detail of two valves. SEM. Fig. 4: inside view of valve. SEM. Fig. 5: inside view of valve showing rimoportula internally. SEM. Fig. 6: detail of seta base with poroid areolae and no spines. TEM. Fig. 7: another seta with spines. TEM. Fig. 1 , bar = 20 urn; Figs 2, 4, bar = 10 urn; Figs 3, 5, 7, bar = 14 D. U. HERNANDEZ-BECERRIL Plate 9 Chaetoceros densus. Fig. 1 : inside view of valve. SEM. Fig. 2: valve with central annulus and rimoportula (arrow). TEM. Fig. 3: detail of annulus and rimoportula. TEM. Fig. 4: internal view of rimoportula (left) and seta base with mesh of fine poroids. SEM. Fig. 5: broken seta. SEM. Fig. 6: seta with poroid areolae and spines. TEM. Figs 1 , 2, bar = 5 urn; Figs 3-6, bar = 2 um. Terminal and intercalary setae differ only in orientation. Several chloroplasts present in cell and setae. Measurements: 31-36 um a.a., 28-36 um p.a., 6-9 um ap. (PI. 10, Figs 1 , 2) EM. This species has the general pattern of a perforated valve without apparent costae. The valve face is reduced, conical, and with an excentric annulus (PI. 10, Figs 3-5, 7). The rimoportula is at the centre of the annulus and has a simple hole inside (with no labiate structure) as well as an external projection, a spine-like structure that is present only in terminal valves (PI. 10, Figs 5, 7). The setae bases are circular in cross-section and very wide, occupying a large area of the valve face. Sibling setae are fused at the point where a tooth-like structure occurs, which suggests that this structure is used for forming chains (PI. 10, Fig. 4). The morphology of the terminal setae does not appear to vary. Although spines are lacking at the base of the setae, they have been shown to be present at distal parts. The wall has a poroid areola pattern, with one stria between two costae (PI. 10, Fig. 6). DISTRIBUTION. 65. REMARKS. Although two forms of this species have been described or illustrated, as well as the separate entity C. STUDY OF CHAETOCEROS SPECIES 15 Plate 10 Chaetoceros denticulatus. Figs 1 , 2: part of chain and two sibling valves, respectively. LM. Fig. 3: two sibling valves. SEM. Fig. 4: detail of tooth-like structure in the setae. SEM. Fig. 5: another valve showing rimoportula and tooth-like structures in the setae. SEM. Fig. 6: base of setae. SEM. Fig. 7: internal view of rimoportula (arrow). SEM. Fig. l,bar = 20 urn; Figs 2, 3, bar = 10 urn; Figs 4, 5, 7, bar = Sum; Fig. 6, bar = 2 urn. 16 nanodenticulatus, it is reasonable to think that C. denticulatus presents a wide range of variation and that those taxa should be considered synonymous, as claimed by Simonsen (1974). It is interesting to note that C. denticulatus forms chains using a tooth-like structure. This character has not previously been reported within Chaetoceros. 9. Chaetoceros rostratus Lauder Lauder, 1864: 79, pi. 8, fig. 10; Hustedt, 1921: pi. 337, fig. 8; Allen & Cupp, 1935: 136, fig. 55; Leger, 1973: 14, figs 2, 3; Navarro, 1982: 316, fig. 40; Giuffre & Ragusa, 1988: 504, figs 1-22. Chaetoceros glandazii L. Mangin Mangin, 1910: 346, fig. 2; Hendey, 1964: 121, pi. 9, fig. 4. Chaetoceros rostratus var. glandazii (L. Mangin) F. J.R. Taylor Taylor, 1966: 439, pi. 2, fig. 12. Pis 11, 12. LM. Chains straight, usually short. In girdle view, cells hexagonal with a process in the centre of the valve connecting sibling cells, corners smooth; sibling cells not touching at the corners, but as setae diverge they leave apparent apertures, which may be narrow or wide. Valve face conical, mantle high, lines of girdle straight. In valve view, cells nearly circular to elliptical. Setae thick, spine-bearing, arising close to the corners but diverging almost perpendicularly to the chain axis, meeting sibling setae, apparently without fusing. Several chloroplasts present in cell and setae. Measurements: 9-20 um a.a., 22-34 urn p.a. (PI. 11, Figs 1,2) EM. Valves are thickly silicified and regularly perforated by poroids. Some have a costa pattern radiating from the centre (PI. 11, Fig. 5; PI. 12, Fig. 3), while in others these costae are either very weak or non-existent (PI. 11, Fig. 4). The poroids become denser in the mantle, arranged in rows between costae running parallel to the pervalvar axis. The processes connecting sibling cells are solid and may be short and thick or longer and thinner (PI. 11, Figs 3-6; PI. 12, Figs 2-4); terminal valves lack the linkage process (PI. 12, Fig. 1). To one side of these processes there is a small, round rimoportula (PI. 12, Figs 2, 4). In the same intercalary cell of the chain the two valves are twisted, so the setae arise without meeting the corresponding sibling setae (PI. 11, Fig. 6; PI. 12, Fig. 2). The seta structure is similar to that described earlier for C. atlanticus var. neapolitanus (p. 4); setae are four-sided with rows of spines along each edge, an areolate, poroid wall, and two stria between pairs of costae (PI. 12, Figs 6, 7)- DISTRIBUTION. 51,65. REMARKS. Controversy has arisen concerning the identity of C. glandazii and its relationship to C. rostratus (e.g. Giuffre & Ragusa, 1988). These authors believe that C. glandazii is a good species. However, the arguments for maintaining it as a separate taxon seem confused. Although no type material has been studied, I concur with authors such as Hustedt (1930) in placing C. glandazii as a synonym of C. rostratus, unless further studies indicate otherwise. Another interesting point has been mentioned by Rines & Hargraves (1988) and concerns the manner in which C. rostratus forms chains through linking processes and the taxonomic significance that this holds. Rines & Hargraves noted that in this character C. rostratus differs from the many other species in section Borealia, but any taxonomic changes are pending, awaiting a major revision of the genus. Whether solitary cells (PI. D. U. HERNANDEZ-BECERRIL 12, Fig. 5) should be considered as distinct forms is better left for future research as information from culture studies is required. 10. Chaetoceros danicus Cleve Cleve, 1889: 55; Peragallo & Peragallo, 1897-1908: 479, pi. 127, fig. 1; Hustedt, 1930: 659, fig. 373; Lebour, 1930: 124, fig. 89; Gran & Angst, 1931: 470, fig. 51; Cupp, 1943: 109, fig. 65; Cleve-Euler, 1951: 96, figs 185a-d; Okuno, 1956: 187, pi. 3, figs 1-5; Hendey, 1964: 122, pi. 10, fig. 5; Brunei, 1970: 88, pi. 11, figs 1-6; Okuno, 1970: pi. 651; Drebes, 1974: 66, fig. 51; Koch & Rivera, 1984: 71, figs 53-56; Rines & Hargraves, 1988: 49, figs 95-99. Pis 13, 14. LM. Cells solitary or united to form very short chains, 2-3 cells long. In girdle view, cells rectangular with small constrictions at girdle level. Valve face flat, with a small process in the centre, mantle high, lines of girdle straight; apertures very narrow, when present. In valve view, cells elliptical, the setae diverging at an angle of about 15° from the apical axis. Setae straight, rather thick, bearing spines, arising from the comers almost parallel to the apical axis. Numerous small chloroplasts present in cell and setae. Measurements: 15-22 um a.a., 16-28 um p.a. (PI. 13, Figs 1,2) EM. The valve is perforated by poroids but lacks costae. The poroids become fewer around the centre of the valve and at the base of the setae; the valve appears thicker at these parts (PI. 13, Figs 5-7; PI. 14, Figs 2, 3). The annulus is inconspicuous or appears without clear boundaries (PI. 13, Fig. 5). Thickenings are present in the mantle, in rows parallel to the lines of the girdle (PI. 14, Fig. 1). The single rimoportula is central, oblong or slit-like, with a small projection outside and a simple hollow inside (PI. 13, Fig. 7; PL 14, Figs 2-4). The setae are circular in cross-section at the base, without spines or areolae, but further from the base they become four-sided, with rows of spines on the edges and single rows of poroids between costae (PI. 14, Figs 5-7). Close to the tip the spines become larger; the tip itself is rather flattened (PI. 14, Fig. 8). In chains, the sibling setae fuse above a short, free base and then diverge again more widely (PI. 13, Fig. 6). DISTRIBUTION. 17,42. REMARKS. This is a very well-known species which has been widely studied (e.g. Okuno, 1956, 1970; Koch & Rivera, 1984). 3. Section PERUVIANA Hern.-Bec., sectio nova Cellulae in catenas vel solitariae. Valvae in cellula, dissimile. Setae robustae, in planitie catena divergentes, versus idem terminus. Rimoportula in uterque valva, lateralis valvaris. Cells solitary or in chains, heterovalvar. All setae robust, pointed towards the same end. Rimoportula present in every valve, excentrically placed. The first attempt to group the species now placed in this section was made by Mangin (1919), who suggested a number of criteria on which to base the inclusion of taxa. The species included are: Chaetoceros aequatorialis Cleve, C. concavicornis L. Mangin, C. convolutus Castrac., C. criophilus Castrac., C. curvatus Castrac., C, pendulus G. Karst., and C. peruvianus Brightw. STUDY OF CHAETOCEROS SPECIES 17 Plate 1 1 Chaetoceros rostratus. Figs 1 , 2: typical chains. LM. Fig. 3: part of chain. SEM. Fig. 4: two sibling valves with short linking process. TEM. Fig. 5: valve with longer linking process. TEM. Fig. 6: sibling valves twisted in relation to each other. SEM. Figs 1 , 2, bar = 20 urn; Figs 3, 6, bar - 10 |im; Figs 4, 5, bar = 5 |im. 18 D. U. HERNANDEZ-BECERRIL Plate 12 Chaetoceros rostratus. Fig. 1 : terminal valve without linking process. SEM. Fig. 2: part of chain with inside view. SEM. Fig. 3: complete cell. TEM. Fig. 4: valve showing rimoportula (arrow). SEM. Fig. 5: single cell (solitary?). SEM. Fig. 6: square (in cross-section) seta with rows of spines. SEM. Fig. 7. Base of seta lacking spines but showing poroid areolae. TEM. Figs 1 , 3, 5, bar = 10 um; Figs 2, 4, bar = 5 um; Figs 6, 7, bar = 2 urn. STUDY OF CHAETOCEROS SPECIES 19 Plate 13 Chaetoceros danicus. Figs 1 , 2: valves in girdle and valve view, respectively. LM. Fig. 3: complete cell. SEM. Fig. 4: valves of two sibling cells. SEM. Fig. 5: valve showing presence of rimoportula. TEM. Fig. 6: sibling valves. SEM. Fig. 7: another single valve with rimoportula (arrow). SEM. Figs 1,2, bar = 20 urn; Figs 3, 4, bar = 10 urn; Figs 5-7, bar = 5 urn. 20 D. U. HERNANDEZ-BECERRIL Plate 14 Chaetoceros danicus. Fig. 1 : valve with thickenings and rimoportula. SEM. Fig. 2: two valves with rimoportula but no apparent annulus. TEM. Fig. 3: two valves with thickenings. TEM. Fig. 4: detail of rimoportula internally (arrow). SEM. Fig. 5: middle part of seta with spines. SEM. Fig. 6: base and distal part of seta. TEM. Fig. 7: another view of middle part of seta. TEM. Fig. 8: tip of seta. TEM. Figs 1, 2, 5, 7, 8, bar = 2 urn; Figs 3, 4, 6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 21 11. Chaetoceros concavicornis L. Mangin Mangin, 1917: 704, figs 5(1), 6, 7; Hustedt, 1930: 665, fig. 376; Gran & Angst, 1931: 469, fig. 50; Cupp, 1943: 109, figs 66a-c; Hendey, 1964: 122, pi. 9, fig. 1; Brunei, 1970: 90, pi. 12, figs 1-5; Evensen & Hasle, 1975: 158, figs 15-22. Chaetoceros criophilus Gran, non C. criophilus Castrac. sensu Gran, 1905: 71, fig. 85. Chaetoceros concaviformis (sic) L. Mangin sensu A. Cleve, 1951 : 96,figsl82a-c. PI. 15, Figs 4, 5. LM. Chains straight, rather short (3-9 cells), relatively robust. In girdle view, cells cylindrical, heterovalvar, the anterior valve rounded, the posterior valve more rectangular. Apertures very small, elongated (in the chain axis). In valve view, cells elliptical. Setae of the anterior valve arise from near the centre of the valve face and diverge; setae of the posterior valve are borne close to the corners and diverge. Distinct terminal setae apparently not present. Many small, rounded chloroplasts present in the cell (also reported within the setae). Measurements: 15-16 um a.a., 35-39 um p.a., 4-5 um ap. (PI. 1 5, Fig. 4) EM. One posterior valve was observed by SEM and seems to be heavily silicified. No distinct valve face was noticed as the bases of the setae are placed closely and are slightly raised (PI. 1 5, Fig. 5). The most outstanding character is the presence of 'prehensors' close to the setae bases on the posterior valve (PI. 15, Fig. 5); these structures should hold the sibling setae, as described for C. convolutus Castrac. (see p. 22). No further details of the setae were observed. DISTRIBUTION. 11, 15. REMARKS. This species has been studied previously using EM (Evensen & Hasle, 1975); their details of the setae (e.g. structure of the wall, tip, etc.) augment the observations made here, although they did not report the presence of prehensors. 12. Chaetoceros convolutus Castrac. Castracane, 1886: 78; Hustedt, 1930: 668, fig. 378; Gran & Angst, 1931: 468, fig. 49; Hendey, 1937: 293; Cupp, 1943: 1 10, fig. 67; Cleve-Euler, 1951: 96, figs 183a-c; Hendey, 1964: 122, pi. 8, fig. 1; Brunei, 1970: 91, pi. 13, figs 1-3; Fryxell & Medlin, 1981: 9, figs 6-8, 43^8; Rines & Hargraves, 1988: 47, fig. 94. PI. 15, Figs 1-3. LM. Chains straight, mostly short (2-8 cells), robust. In girdle view, cells cylindrical, heterovalvar, the anterior valve more Plate 15 Chaetoceros convolutus (Figs 1-3) and C. concavicornis (Figs 4, 5). Fig. 1 : broken chain with two valves. LM. Fig. 2: two sibling valves in chain. SEM. Fig. 3: detail of valves showing prehensor holding sibling seta. SEM. Fig. 4: typical chain. LM. Fig. 5: valve showing prehensors on setae. SEM. Figs 1 , 4, bar = 50 um; Fig. 2, bar = 20 um; Figs 3, 5, bar = 10 um. 22 D. U. HERNANDEZ-BECERRIL rounded than the rectangular posterior valve. Apertures very narrow or almost absent; girdle zone very conspicuous. In valve view, cells subcircular or elliptical. Setae becoming thicker and spine-bearing at distal parts; those of the anterior valve arising close to the centre and diverging widely; those of the posterior valve arising from close to the corners and spreading. Numerous chloroplasts present in the cell and setae. Measurements: 22-42 urn a.a., 28-34 urn p.a. (PI. 15, Fig. 1) EM. Two sibling intercalary valves were observed (PI. 15, Fig. 2). They appear to be heavily silicified and perforated by small, randomly distributed poroids (PI. 15, Fig. 3). The rounded anterior valve shows no clear valve face and the setae bases occupy most of the centre of this area. An external protrusion of the rimoportula was noticed on one side of the posterior valve (the rectangular one) (PI. 15, Fig. 3); no further observations were made on the corresponding rimoportula of the anterior valve. There is no apparent fusion of the sibling setae, but at the setae bases on the posterior valve some sort of prehensors develop (one per seta) and hold the setae arising from the sibling anterior valve (PI. 15, Fig. 3). Sibling setae then diverge opposite each other. The setae become thicker and four-sided distally, with rows of spines at the edges (PI. 15, Fig. 3). DISTRIBUTION. 12. REMARKS. Chaetoceros concavicornis and C. convolutus show a particular type of chain-formation, namely the possession of prehensors to hold sibling setae in the chain, which was first described by Fryxell & Medlin (1981), at least for C. convolutus. Fryxell & Medlin (1981) also discussed the difficulty of distinguishing between these two species. 13. Chaetoceros peruvianus Brightw. Brightwell, 1856: 107, pi. 7, figs 16-18; Hustedt, 1930: 671, fig. 379, 380; Allen & Cupp, 1935: 136, fig. 56; Hendey, 1937: 296, pi. 13, fig. 6; Cupp, 1943: 113, fig. 68a-c; Okuno, 1956: 191, pi. 8, fig. 1-8; Hendey, 1964: 123, pi. 9, fig. 3; Okuno, 1970: pis 655, 656; Koch & Rivera, 1984: 69, figs 36-47; Desikachary et al., 1987: pi. 354, fig. 4; Rines & Hargraves, 1988: 53, figs 108-112. Chaetoceros peruvianus f. robusta (Cleve) Hust. Hustedt, 1930, p. 673, fig. 38 la. Chaetoceros peruvianus f. gracilis (Schrod.) Hust. Hustedt, 1930: 672, fig. 381 b. Chaetoceros chilensis Krasske Krasske, 1941: 266, pi. 4, fig. 3. Pis 16, 17. LM. Cells solitary, robust, heterovalvar. In girdle view, cells cylindrical, with rounded corners and constrictions at girdle zones. Anterior valve face reduced, mostly occupied by the setae bases, mantle variable in depth, lines of girdle straight; posterior valve face flat. In valve view, cells elliptical. Setae thick, coarse, long, bearing spines; setae of anterior valves fused together after a very short, free base, then diverging widely towards the pervalvar axis; setae of posterior valves arising close to the corners of the valve, then also diverging towards the pervalvar axis. Small chloroplasts present in cell and setae. Measurements: 13-22 urn a.a., 20-36 urn p.a. (PI. 16, Fig. 1) EM. The valves are heavily silicified (PI. 16, Fig. 2) and perforated by poroids. In the centre of some anterior valves a pattern of weak costae runs parallel to the transapical axis (PI. 16, Figs 3-6). No costae were seen in posterior valves (PI. 17, Fig. 3). The annulus is oblong, placed to one side of the valve, and the rimoportula occurs at its centre (PI. 17, Fig. 2). The rimoportula is a simple hole inside but projects externally as a flattened tube (PI. 16, Fig. 4); in posterior valves the rimoportula appears to be open to the outside and it is relatively shorter and more cylindrical (PI. 17, Fig. 3). The rimoportulae in both valves are apparently located to the same side. Anterior setae fuse together over the valve centre, forming an interlocking tooth-like structure (PI. 17, Fig. 1), before twisting slightly and showing the common structure, as mentioned for most of the species within subgenus Chaetoceros. The setae are marked with longitudinal ridges at the base and are generally four-sided; spines are also present in rows and there are three striae between costae (PI. 17, Figs 5-8). DISTRIBUTION. 3, 6, 7, 10, 15-18, 23, 24, 26, 36, 42, 45, 48, 58, 64. REMARKS. Previous EM observations are confirmed here. However, despite Koch & Rivera's (1984) statements about the variation in the number of striae between costae, it is clear from their figures and my own observations that the areola pattern shows no change in this aspect; Koch & Rivera's (1984) interpretations were based on setae observed at different angles, such that it appears the intercostal striae may number more than three. Ratkova (1974) tried to show that differences exist between C. chilensis and C. peruvianus, but her observations do not seem to be conclusive. 14. Chaetoceros pendulus G. Karst. Karsten, 1905: 1 18, pi. 15, figs 7, 7a; Hendey, 1937: 295; Cupp, 1943: 114, fig. 69. Pis 18, 19. LM. Cells solitary, heterovalvar. In girdle view, cells almost rectangular, with smooth corners and constricted at the girdle zones, the apical axis being relatively longer. Anterior valve convex, mantle very low, lines of girdle straight; posterior valve depressed at centre. In valve view, cells elliptical. Setae thick, long, bearing spines distally; those of anterior valve arise diagonally, then curve to the pervalvar axis, directed almost parallel to the setae from the posterior valve, which simply curve smoothly towards the pervalvar axis. Several small chloroplasts present in cell and setae. Measurements: 15-19 um a.a., 10-17 urn p.a. (PI. 18, Figs 1,2) EM. The valves are fairly heavily silicified (PI. 18, Figs 3, 4, 6), perforated by small poroids (PI. 18, Fig. 5), and show a pattern of weak costae arranged in rows following the apical axis close to the base of the setae (PI. 19, Fig. 2). A small number of slightly stronger costae are present in the centre of the valve, running transapically and leading to the excentric annulus. In the centre of this annulus a rimoportula is found, comprising a simple oblong hole internally, with a connection close to it, and a very short, flattened tube projecting externally (PI. 19, Figs 1-3). The rimoportulae of both the posterior and the anterior valves are at the same side of the cell, the one in the posterior valve located in a depressed area of the valve (PI. 18, Figs 4, 5). The setae bases are circular and simply perforated. Setae become four-sided, with large spines running in rows along each edge; their walls are poroid, with three striae between costae (PI. 19, Figs 4-7). The tips are very sharp, without any opening to the outside (PI. 19, Fig. 8). DISTRIBUTION. 24, 48, 50. STUDY OF CHAETOCEROS SPECIES 23 Plate 16 Chaetocerosperuvianus. Fig. 1 : complete cell. LM. Fig. 2: complete cell. SEM. Fig. 3: another cell with various girdle bands. SEM. Fig. 4: anterior valve showing rimoportula (arrow). SEM. Fig. 5: another partially broken anterior valve. TEM. Fig. 6: detail of posterior valve. SEM. Figs 1 , 2, bar = 20 \un; Figs 3-6, bar = 5 urn. 24 D. U. HERNANDEZ-BECERRIL Plate 17 Chaetocerospemvianus. Fig. 1 : anterior valve showing interlocking of setae. SEM. Fig. 2: anterior valve with excentric annulus and rimoportula (arrow). TEM. Fig. 3: posterior valve with rimoportula (arrow). TEM. Fig. 4: internal view of rimoportula (arrow). SEM. Fig. 5: square (in cross-section) seta. SEM. Figs 6, 9: middle part of seta with spines. SEM. Figs 7, 8: middle part of seta. TEM. Figs 1 , 2, 4, 5, 9, bar = 5 jim; Fig. 3, bar = 10 \un; Figs 6-8, bar = 2 ^m. STUDY OF CHAETOCEROS SPECIES 25 Plate 18 Chaetocerospendulus. Figs 1,2: solitary cells. LM. Fig. 3: complete cell. SEM. Fig. 4: detail of cell showing rimoportula (arrow) at poste- rior valve. SEM. Fig. 5: valve with setae bases. TEM. Fig. 6: anterior and posterior valves separated. SEM. Figs 1 , 3, bar = 50 \un; Figs 2, 6, bar = 10 urn; Figs 4, 5, bar = 5 urn. 26 D. U. HERNANDEZ-BECERRIL Plate 19 Chaetoceros pendulus. Fig. 1 : anterior valve showing excentric rimoportula. SEM . Fig. 2: another anterior valve with excentric annulus and rimoportula. TEM. Fig. 3: inside view of valve. Arrow indicates rimoportula. SEM. Fig. 4: base of seta. TEM. Fig. 5: four-sided (in cross- section) seta. SEM. Fig. 6: middle part of seta. SEM. Fig. 7: middle part of seta with spines. TEM. Fig. 8: tip of seta. SEM. Figs 1-3, bar = 5 um; Figs 4-8, bar = 2 um. STUDY OF CHAETOCEROS SPECIES 27 REMARKS. This species is closely related to C. aequatorialis Cleve and further studies may reveal that they are conspecific. Cleve (18730: pi. 2, fig. 9) figured one specimen forming a short chain (two cells), accompanied by a rather poor description. The species described here is identified as C. pendulus since this has previously been reported from the area where the present material was collected (Cupp, 1 943) and because chains were not found in this material. However, if both species are indeed conspecific, the name C. aequatorialis would have priority. 4. Section COARCTATA Hern.-Bec. Chains heteropolar, usually very robust. Numerous small rimoportulae present in each valve. 15. Chaetoceros coarctatus Lauder Hernandez-Becerril, 1991a: 282, figs 1-13. DISTRIBUTION. 7, 16, 17, 19,20,23,41,44,45,48,49. II. Subgenus HYALOCHAETE Gran 5. Section DICLADIA (Ehrenb.) Gran Terminal and intercalary setae similar. Chloroplasts numerous. Some authors (Lebour, 1930; Cupp, 1943; Hendey, 1964) have included C. decipiens Cleve within section Oceanica Ostenf. because it lacks resting spores. However, I consider that C. decipiens and C. lorenzianus Grunov are closely related, and I include the former in this section. 1 6. Chaetoceros decipiens Cleve Cleve, 18736: 11, pi. 1, figs 5a, b; Peragallo & Peragallo, 1897-1908: 485, pi. 131, figs 3, 4, 8; Hustedt, 1930: 675, fig. 383; Gran & Angst, 1931: 470, fig. 52; Cupp, 1943: 115, figs 70_A, 70-Ba, b; Cleve-Euler, 1951: 98, figs 188a, b; Hendey et al., 1954: 29, pi. 1, fig. 4; Okuno, 1956: 191, pi. 7, figs 1-8; Hendey, 1964: 123, pi. 12, fig. 2; Brunei, 1970: 99, pi. 21, figs 1^, pi. 22, figs 1-5; Drebes, 1974: 69, fig. 52; Evensen & Hasle, 1975: 161, figs 55-69; Navarro, 1982: 309, fig. 22; Rines & Hargraves, 1988: 75, figs 148, 152. Pis 20, 21. LM. Chains straight, often long. In girdle view, cells rectangular, with sharp corners. Valve face nearly flat or slightly concave, mantle high, lines of girdle straight; apertures rather wide. In valve view, cells elliptical, the setae straight, following the apical axis. Setae relatively thick, with minute striae along their length; terminal setae slightly thicker than intercalary setae and also differing in their direction. Intercalary setae arise from the corners and fuse immediately with sibling setae, remaining joined for a short distance before diverging again at an angle of about 10-15° from the apical axis. Small chloroplasts present at the centre of the cell. Measurements: 18-40 um a.a., 17-24 um p.a., 5-10 um ap. (PI. 20, Figs 1, 2) EM. The valve morphology of this species may be considered as typical for most of the species within subgenus Hyalochaete: the valves do not appear to be as heavily silicified as those of species placed in subgenus Chaetoceros (PI. 20, Figs 3, 4; PI. 21, Figs 3, 4). Another general characteristic of the valves is the presence of an annulus from which a weak costal pattern radiates (PI. 20, Figs 4, 5). There are small poroids between the costae. A hyaline rim is often found on the edge of some terminal valves (PI. 21, Fig. 5). Also, short, hair-like filaments of solid appearence become visible in apical areas of some intercalary valve mantles (PI. 20, Fig. 6; PI. 21 , Fig. 2). Thin walls with a simple, perforated structure appear to connect the corners (apical areas) of intercalary sibling valves (PI. 21, Fig. 2). A rimoportula is present only on terminal valves; each shows a very small projection to the outside and is centrally placed within the annulus (PI. 21, Fig. 5). The setae are circular at the base (PI. 21, Fig. 1), but become polygonal (6-8 sides) distally, with spines running in rows along each edge and rather large perforations in the sides (PI. 21, Figs 2, 6, 7). Terminal setae are thicker than intercalary setae, with relatively bigger spines, but smaller holes (PI. 21, Fig. 6). The tips are truncated (PI. 21, Fig. 7). DISTRIBUTION. 1 5-1 7, 20, 23, 3 1 , 35, 36, 42, 52. REMARKS. See remarks under C. lorenzianus Grunov (p. 3 1 ). 17. Chaetoceros decipiens f. singularis Gran Gran, 1904: 536, pi. 17, fig. 7; Cupp, 1943: 1 17, fig. 70-Bc, d. PI. 24, Figs 1,2. LM & EM. Cells solitary. The morphology is typically the same as that of the type, with setae diverging at an angle of about 10-20° from the pervalvar axis. Valves and setae are as described above. Two snail-like structures were found at the base of the setae (PI. 24, Figs 1,2). DISTRIBUTION. 1 7, 23, 42, 45. REMARKS. It is unclear whether this taxon should be considered as a form of C. decipiens or if it is only part of the life cycle of the species. 18. Chaetoceros lorenzianus Grunov Grunow, 1863: 157, pi. 5, fig. 13; Peragallo & Peragallo, 1 897-1 908: 484, pi. 1 30, figs 1 , 2, pi. 1 3 1 , fig. 1 , non 3; Hustedt, 1930: 679, fig. 385; Gran & Angst, 1931: 471, fig. 53; Allen & Cupp, 1935: 137, fig. 58; Cupp, 1943: 118, fig. 71; Subrahmanyan, 1946: 131, figs 199, 202-204, 206-209, non 198; Cleve-Euler, 1951: 98, figs 187a-d; Okuno, 1956: 190, pi. 6, figs 1-7; Hendey, 1964: 124, pi. 16, fig. 1; Brunei, 1970: 105, pi. 18, figs 1-4, pi. 19, figs 1-4; Navarro, 1982: 312, fig. 34; Desikachary & Prema, 1987: pi. 262, figs 3, 4 (non C. lorenzianus Grunov sensu Helmcke & Krieger, 1954: pi. 124 = C. decipiens Cleve). Pis 22, 23. LM. Chains straight, of variable length. In girdle view, cells rectangular with sharp comers, the pervalvar axis generally longer. Valve face flat or slightly convex in the centre, mantle high, lines of girdle straight; apertures variable in size. In valve view, cells elliptical. Setae relatively thick, marked with striae; intercalary setae arise from the valve corners, diverging at an angle of about 15-20° from the apical axis; terminal setae curve towards the chain axis. Several, small chloroplasts present in the cell. Measurements: 13-25 um a.a., 8-50 urn p.a., 8-12 um ap. (PI. 22, Fig. 1) EM. The valve structure is similar to that in C. decipiens (PI. 22, Fig. 2), but the costae are more marked and stronger (PI. 22, Figs 3-6). The costae converge at the apices, making the valve thicker. Small poroids are scattered between the costae. The girdle structure differs slightly from that of the valve in lacking costae, 28 D. U. HERNANDEZ-BECERRIL Plate 20 Chaetoceros decipiens. Figs 1,2: complete chain and part of chain, respectively. LM. Fig. 3: part of chain. SEM. Fig. 4: sibling valves, the lower showing some filaments (arrows). TEM. Fig. 5. valve view of valve with central annulus. TEM. Fig. 6: sibling cells, one showing filaments in mantle (arrow). SEM. Fig. 1 , bar = 100 urn; Figs 2, 3, bar = 20 \un; Figs 4-6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 29 Plate 21 Chaetoceros decipiens. Fig. 1 : base of setae. SEM. Fig. 2: base of setae joined for a short distance. Arrow indicates filaments on mantle. TEM. Fig. 3: terminal valve. TEM. Fig. 4: valve view of chain. SEM. Fig. 5: terminal valve with central rimoportula. SEM. Fig. 6: terminal seta (upper) and intercalary setae (lower). SEM. Fig. 7: tip of seta. SEM. Figs 1 , 2, 5, 7, bar = 2 urn; Figs 3, 4, 6, bar = 10 \im. 30 D. U. HERNANDEZ-BECERRIL Plate 22 Chaetoceros lorenzianus. Fig. 1 : typical chain. LM. Fig. 2: complete chain. SEM. Fig. 3: sibling valves. SEM. Fig. 4: valve with several gir- dle bands. TEM. Fig. 5: broken valve showing base of setae. TEM. Fig. 6: terminal valve with rimoportula. TEM. Figs 1 , 2, bar = 20 jim; Figs 3, 4, 6, bar = 5 urn; Fig. 5, bar = 10 \m\. STUDY OF CHAETOCEROS SPECIES 31 Plate 23 Chaetoceros lorenzianus. Fig. 1 : terminal cell with rimoportula at terminal valve (arrow). SEM. Fig. 2: terminal valve with tube-like rimoportula. TEM. Fig. 3: middle part of seta. TEM. Fig. 4: middle part of seta. SEM. Fig. 5: inside view of valve showing rimoportula with labi- ate structure. SEM. Fig. 1, bar = 5 urn; Fig. 2, bar = 10 urn; Figs 3-5, bar = 2 urn. but with rows of poroids that follow the pervalvar axis (PI. 22, Fig. 4). The rimoportulae are found only on terminal valves, located in the centre, again coinciding with the annulus. Each consists of a relatively short or longer cylindrical tube open to the outside and with a true labiate structure inside (PI. 23, Figs 1 , 2, 5). The structure of the setae is very like that of C. decipiens and it is difficult to determine any differences between them (PI. 23, Figs 3, 4). No resting spores were observed. DISTRIBUTION. 12, 15, 18,23,42,45,48, 53. REMARKS. Okuno (1956) studied C. decipiens and C. lorenzianus and illustrated the major differences between them. Apparently, C. lorenzianus presents a stronger costal pattern and lacks the fine poroids which occur in the valves of C. decipiens (Evensen & Hasle, 1975). Furthermore, following the observations made in this study, it can be appreciated that the distinctive structure of the rimoportulae, with that of C. lorenzianus projected externally in a short tube and that of C. decipiens being rather reduced, could be an important distinguishing character, as could the distinctive presence of filaments in C. decipiens, which are absent in C. lorenzianus. The presence of resting spores in C. lorenzianus is a useful character for identification. However, the possibility that both are stages in the life cycle of one species calls for further investigation (see also Evensen & Hasle, 1975). 19. Chaetoceros lorenzianus f. forceps Meunier Meunier, 1913: 20, pi. 2, figs 12-18; Brunei, 1970: 106, pi. 20, figs 1-3. PI. 24, Figs 3-6. LM & EM. This is a solitary, non chain-forming form of C. lorenzianus, which has the same basic characteristics regarding valve and seta morphology (PI. 24, Figs 3-6). No resting spores were found. DISTRIBUTION. 23. REMARKS. Brunei ( 1 970) has shown that this form is indeed able to create resting spores which are identical to those of the type species. 32 D. U. HERNANDEZ-BECERRIL Plate 24 Chaetoceros decipiens f. singularis (Figs 1 , 2) and C. lorenzianus f. forceps (Figs 3-6). Fig. 1 : complete cell. LM. Fig. 2: complete cell. SEM. Fig. 3: complete cell. LM. Fig. 4: detail of seta tip. SEM. Fig. 5: complete cell. SEM. Fig. 6: cell in valve view. LM. Fig. 1 , bar = 50 \an; Fig. 2, bar = 10 urn; Figs 3, 5, 6, bar = 20 )im; Fig. 4, bar = 2 ^m. 6. Section CYLINDRICA Ostenf. Valves nearly circular (cylindrical). Apertures narrow. Terminal setae not thicker than other setae. Chloroplasts numerous. 20. Chaetoceros bermejensis Hern. -Bee. Hernandez-Becerril, 1991c: 522, figs 1-14. DISTRIBUTION. 42,45. 7. Section COMPRESSA Ostenf. Valves compressed. Intercalary setae of two types: thin, common setae and thicker, twisted special setae. 21 . Chaetoceros compressus Lauder Lauder, 1864: 78, pi. 8, fig. 6; Peragallo & Peragallo, 1897-1908: 488, pi. 134, fig. 8; Hustedt, 1930: 684, fig. 388; Gran & Angst, 193 1 : 472, fig. 55; Allen & Cupp, 1935: 1 38, fig. 60; Cupp, 1943: 119, fig. 74; Subrahmanyan, 1946: 134, fig. 218; Cleve-Euler, 1951: 99, figs 191b, c, non a; Hendey et al., STUDY OF CHAETOCEROS SPECIES 33 Plate 25 Chaetoceros compressus. Figs 1 , 2: typical chain and part of chain, respectively. LM. Fig. 3: part of chain with special setae. SEM. Fig. 4: another part of chain. SEM. Fig. 5: sibling valves with special setae. SEM. Fig. 6: middle part of special seta. TEM. Fig. 7: tip of special seta. SEM. Figs 1-4, bar = 20 urn; Fig. 5, bar = 10 urn; Figs 6, 7, bar = 5 jun. 34 1954: 29, pi. 1, fig. 3, non fig. 5; Hendey, 1964: 125, pi. 15, fig. 5; Brunei, 1970: 1 10, pi. 25, figs l^t; Drebes, 1974: 72, figs 56a, b;Navarro, 1982:309,fig. 15;Desikacharyetal., 1987:pl.258, fig. 2; Rines& Margraves, 1988: 64, figs 131-134, 218. Pis 25, 26. LM. Chains variable, straight or slightly twisted, generally long and delicate. In girdle view, cells rectangular to square, with sharp corners. Valve face convex, mantle low, lines of girdle straight; apertures of variable width. In valve view, cells elliptical. Terminal and intercalary setae differ, the former thicker and coarser. Special intercalary setae often present, longer and with larger spines than other intercalary setae. Common intercalary setae diverging at random. Various small chloroplasts present in the cell. Measurements: 10-21 um a.a., 10-24 um p.a., 6-7 um ap. (PI. 25, Figs 1, 2) D. U. HERNANDEZ-BECERRIL EM. The valve is usually weakly silicified (PI. 25, Figs 3, 4), with a conspicuous annulus from which weak costae radiate (PI. 25, Fig. 4; PI. 26, Fig. 1). The rimoportula is a rather excentric, hollow structure and occurs in the annulus (PI. 26, Figs 1 , 2). The setae arise from close to the corners of the valve (PI. 26, Fig. 1); they have a spiral pattern of spines running longitudinaly, as well as small poroids perforating the wall (PI. 25, Fig. 6). The spines on special setae are larger than those on common intercalary setae (PI. 25, Figs 3, 5; PI. 26, Figs 3-5, 7). All setae are circular in cross-section and the tips are very pointed (PI. 25, Fig. 7; PI. 26, Fig. 6). DISTRIBUTION. 1, 7, 17, 23, 52, 53. REMARKS. This species exhibits a wide range of forms during its life cycle, which makes positive identification difficult. Plate 26 Chaetoceros compressus. Fig. 1 : terminal cell with rimoportula (arrow) at terminal valve. SEM. Fig. 2: terminal cell showing rimoportula (arrow). SEM. Fig. 3: detail of special seta. SEM. Fig. 4: middle part of special seta. TEM. Fig. 5: middle part of special seta. SEM. Fig. 6: tip of special seta. TEM. Fig. 7: middle part of intercalary (common) seta. SEM. Figs 1-3, bar = 5 um; Figs 4-7, bar = 2 um. STUDY OF CHAETOCEROS SPECIES 35 8. Section PROTUBERANTIA Ostenf. emend. Hern.-Bec. A large protuberance present in the centre of the valve. Two chloroplasts per cell. In the light of recent observations, mainly on C. protuberans Lauder whose resting spores lack setae (Lechuga-Deveze & Hernandez- Becerril, 1988), this section can no longer be regarded as having resting spores with long setae ('two long horns', Lebour, 1930) 22. Chaetoceros didymus Ehrenb. Hernandez-Becerril, 19916: 290, figs 1-12. DISTRIBUTION. 23, 40, 42, 52, 55, 64. 23. Chaetoceros protuberans Lauder Hernandez-Becerril, 19916: 290, figs 14-25. DISTRIBUTION. 4, 7, 8, 13, 23, 41, 44-46, 48. 9. Section CONSTRICTA Gran Cells constricted. Terminal setae thicker than other setae. Chloroplasts one or two. Resting spores with spines on both valves. 24. Chaetoceros constrictus Gran Gran, 1897: 17, pi. 1, figs 11-13, pi. 3, fig. 42; Peragallo & Peragallo, 1897-1908: 491, pi. 134, fig. 5; Hustedt, 1930: 694, fig. 395; Cupp, 1943: 122, fig. 76; Cleve-Euler, 1951: 100, figs 193a-c; Hendey, 1964: 126, pi. 9, fig. 2; Brunei, 1970: 112, pi. 26, figs 1,2; Rines & Hargraves, 1988: 67, figs 128-130. PI. 29, Fig. 6. LM. Chains straight, relatively short. In girdle view, cells rectangular with sharp corners, showing a conspicuous constriction at girdle zones. Valve face slightly concave, leaving a narrow, lanceolate aperture. Intercalary setae orientated close to the apical axis; terminal setae thicker, directed to the chain axis. Two chloroplasts present in each cell, placed at the valve faces. Measurements: 16-18 urn a.a., 13-15 urn p.a., 2 um ap. (PI. 29, Fig. 6) EM. No observations were made. DISTRIBUTION. 24, 41, 48, 52. REMARKS. See next species. 25. Chaetoceros vanheurcki Gran Gran, 1897: 1 8; Gran & Angst, 193 1:476, fig. 60; Cupp, 1943: 123, fig. 77. PI. 29, Figs 4, 5. LM. Chains straight, of variable length, delicate. In girdle view, cells rectangular, with sharp corners, mantle low. Apertures oblong, narrowing at the centre. Intercalary setae directed close to the apical axis; terminal setae slightly thicker, orientated to the chain axis. Two? chloroplasts present. Measurements: 15-16 um a.a., 9-12 um p.a., 2-3 um ap. (PI. 29, Fig. 4) EM. A few observations made using SEM show valves which are not heavily silicified, with clear constrictions near to the girdle (PI. 29, Fig. 5). There appears to be no rimoportula on intercalary valves and therefore it is assumed that this structure is only located on terminal valves. The intercalary setae arise from the very apex of the valves and fuse immediately to the corresponding sibling valve (PI. 29, Fig. 5). DISTRIBUTION. 12,48. REMARKS. The two species studied within section Constricta have traditionally been separated by the morphology of the resting spores. The resting spores of C. vanheurckii have long spines present on the secondary valve, in contrast to C. constrictus, which has short-spined resting spores. This is another case where cultures and life cycle studies are required to show whether or not morphological variation is present in these two species, and to define the important morphological characters used to distinguish between them. The direction of the intercalary setae (e.g. in C. constrictus they diverge more widely than in C. vanheurckii) may not be a character of sufficient taxonomic value to separate them. 10. Section STENOCINCTA Ostenf. Terminal setae curved, thicker than other setae. One chloroplast. Resting spores with numerous spines or processes on both valves. 26. Chaetoceros af finis Lauder var. affinis Lauder, 1864: 78, pi. 8, fig. 5; Peragallo & Peragallo, 1897-1908: 478, pi. 129, fig. 3, pi. 130, fig. 3; Hustedt, 1930: 695, fig. 396; Gran & Angst, 1931: 479, fig. 63; Cupp, 1943: 125, figs 78-A(l), 78-A(2); Cleve-Euler, 1951: 105, fig. 212; Hendey, 1964: 127, pi. 18, fig. 2; Okuno, 1956: 186, pi. 1, figs 1-5; Hendey, 1964: 127, pi. 18, fig. 3; Brunei, 1970: 1 14, pi. 27, fig. 1; Drebes, 1974: 75, fig. 58a; Evensen & Hasle, 1975: 161, figs 46-54; Rines & Hargraves, 1988: 59, figs 1 13, 1 14. The numerous synonyms are listed by Hustedt (1930) and VanLandingham (1968). Pis 27, 28. LM. Chains straight, usually long, robust. In girdle view, cells rectangular, slightly constricted at girdle zones. Valve face concave, mantle high, lines of girdle straight; apertures narrow. In valve view, cells elliptical. Terminal and intercalary setae differ: terminal setae thicker, smoothly curved; intercalary setae diverging at an angle of about 1 5-30° from the apical axis. One chloroplast present at the centre of each cell. Measurements: 18-25 um a.a., 10-22 um p.a., 4-6 um ap. (PI. 27, Fig. 1) EM. Short chains (PI. 27, Fig. 3) and loose valves were observed. The valves are irregularly perforated by poroids and some show a weak costal pattern radiating from the centre (PI. 27, Figs 2, 4, 5, 7). The rimoportula is centrally located, comprising a flattened, tube-like structure externally with a true labiate structure internally (PI. 28, Figs 1, 2). The morphology of the terminal and intercalary setae differ. The terminal setae bear regular rows of large spines which run parallel to their main axis, with rows of poroids also running along the same axis, as well as a scattering of relatively larger poroids. These setae are polygonal in cross-section (PI. 28, Figs 3-5). Intercalary setae have rows of smaller spines and poroids in a spiral pattern, and are rather circular in cross-section (PI. 27, Fig. 6; PI. 28, Figs 6, 7)- DISTRIBUTION. 1 8, 23, 24, 42, 45, 64. REMARKS. The morphology, as described here, is similar to that recorded by Evensen & Hasle (1975) for C. affinis from 36 D. U. HERNANDEZ-BECERRIL Plate 27 Chaetoceros affmis var. affinis. Fig. 1 : typical chain. LM. Fig. 2: single terminal valve. TEM. Fig. 3: part of chain. Note difference between terminal and intercalary setae. SEM. Fig. 4: terminal valve showing rimoportula. SEM. Fig. 5: terminal valve with rimoportula. TEM. Fig. 6: mid- dle part of intercalary seta. SEM. Fig. 7: valve face (terminal valve) with rimoportula. SEM. Figs 1 , 3, bar = 50 urn; Figs 2, 4, 5, bar = 5 \an\ Figs 6, 7, bar = 2 urn. STUDY OF CHAETOCEROS SPECIES Plate 28 Chaetoceros affmis var. affinis. Fig. 1 : terminal valve with external view of rimoportula. SEM . Fig. 2: internal view of rimoportula. Note labiate structure. SEM. Fig. 3: complete terminal seta. SEM. Fig. 4: middle part of terminal seta. TEM. Fig. 5: terminal seta (above) and interca- lary seta (below). SEM. Fig. 6: middle part of intercalary seta. TEM. Fig. 7: intercalary seta close to tip. SEM. Figs 1 , 2, 6, 7, bar = 2 urn; Figs 3-5, bar = 5 urn. 38 D. U. HERNANDEZ-BECERRIL Californian cultures. They claimed to have found at least two taxa included in C. affinis, as currently delimited. 27. Chaetoceros affinis var. willei (Gran) Hust. Hustedt, 1930: 697, fig. 398; Cupp, 1943: 126, fig. 78c; Cleve-Euler, 1951: 105, fig. 213; Brunei, 1970: 116, pi. 27, fig. 2;Drebes, 1974:75,fig.58b. Chaetoceros willei Gran Gran, 1897: 19, pi. 4, fig. 47; Hendey, 1964: 126, pi. 18, fig. 4. PI. 29, Figs 1-3. LM. Chains straight, short, delicate. In girdle view, cells rectangular, with sharp corners, mantle high; apertures very narrow. In valve view, cells elliptical. Terminal setae thicker than intercalary setae, lying c. 25-30° from the chain axis. Intercalary setae arise either parallel to the apical axis or at an angle of 1 5-25° from this axis. One chloroplast present at the centre of the cell. Measurements: 10-12 urn a.a., 17 um p.a. (PI. 29, Fig. 1) EM. Valves not heavily silicified (PI. 29, Fig. 2). No details of the costae pattern were observed. The rimoportula is not prominent (as in var. affinis) and is presumably located only on terminal valves. The terminal setae are slightly thinner than those of var. affinis and have spiral rows of minute spines (PI. 29, Fig. 3). DISTRIBUTION. 23. REMARKS. It is shown here that the morphology of var. willei is clearly similar to that of the type (var. affinis). This is the reason for considering the former species C. willei as a variety of C. affinis. However, further studies, especially by EM, are required for those varieties (and synomyms) of C. affinis which are not well known (e.g. var. circinalis (Meunier) Hust., C. schuetti Cleve, etc.). Observations on the resting spores may show significant differences that warrant separation into separate species. 28. Chaetoceros paradoxus Cleve Hernandez-Becerril, 19936: 170, figs 1-14. DISTRIBUTION. 17,23,24,42,50,52. 1 1 . Section LACINIOSA Ostenf. Apertures wide. Chloroplasts one or two. 29. Chaetoceros distans Cleve Cleve, 1873a: 9, pi. 2, figs 1 la, b; Gran & Yendo, 1914: 9, fig. 10; Hustedt, 1921: pi. 337, figs 3-6, pi. 338, fig. 8 (non Chaetoceros distans var. sensu Cleve, 1894: 14, pi. 2, fig. 2 = ? C. laciniosus F. Schiitt). Pis 30, 31. LM. Chains straight, of variable length. In girdle view, cells rectangular, with sharp corners. Valve face flat or slightly convex, mantle low, lines of girdle curved (convex); apertures wide. In valve view, cells elliptical or nearly circular. Intercalary setae usually thin, arising from the corners, then diverging almost parallel to the apical axis or curving smoothly towards the chain axis. Terminal setae arise straight, then curve to the chain axis. One chloroplast present at the centre of the cell. Measurements: 20-30 um a.a., 30-37 um p.a., 18-24 um ap. (PI. 30. Fig. 1) EM. The chain is delicate (PI. 30, Fig. 2) and the valves are thinly silicified (PI. 30, Fig. 5), having the typical pattern of a conspicuous central annulus and radiating costae (PI. 30, Figs 3, 4, 6), with some poroids between the costae (PI. 31, Figs 1, 4). The costae join together at apical zones close to the base of the setae, making the valve thicker at this point (PI. 39, Figs 3, 6). A hyaline rim appears on the edge of some valves (PI. 30, Figs 4, 6). The rimoportula is a flattened hollow at the centre of the annulus (PI. 31, Figs 3, 4). The intercalary setae have a rather long base (PI. 31, Fig. 2), being circular throughout, with small spines in spiral rows and poroids distributed in the wall with no apparent pattern (PI. 31, Figs 5-8). Terminal setae differ in direction. DISTRIBUTION. 7, 15, 23, 28, 42, 48, 58, 64. REMARKS. The question of whether or not this species is conspecific with C. laciniosus F. Schiitt cannot be answered here, because studies of the life cycle are needed. The general structure appears to be the same (see Evensen & Hasle, 1975 for observations on C. laciniosus). One distinctive character is found in the girdle insertion; the lines of girdle in C. distans are curved, whereas in C. laciniosus they are straight (Gran & Yendo, 1914). Ikari (1928) indicated that 'the basal part' of the setae is shorter in C. laciniosus than in C. distans. These characters may not be sufficient to separate them, and in that case the name C. distans would have priority over C. laciniosus. VanLandingham (1968) cited C. distans as a synonym of C. dichaeta. 30. Chaetoceros brevis F. Schiitt Schiitt, 1895: 38, pis 4, 5, figs 4a, b; Hustedt, 1930: 707, figs 403a,b; Allen & Cupp, 1935: 141, fig. 70; Cupp, 1943: 129, fig. 82; Cleve-Euler, 1951: 103, figs 205a-c, non d; Hendey, 1964: 127, pi. 9, fig. 5; Brunei, 1970: 120, pi. 27, fig. 3, non pi. 29, figs 1,2; Navarro, 1982: 309, fig. 12. Pis 32, 33. LM. Chains straight, generally short and delicate. In girdle view, cells rectangular or square, with sharp corners. Valve face convex, with a slight protuberance, mantle high, lines of girdle straight; apertures wide. In valve view, cells elliptical, with setae diverging only slightly from the apical axis. Setae thin, delicate, arising from the corners of the valves. Intercalary setae diverge almost parallel to the apical axis at an angle of about 15°, or curve towards the main axis, as do the terminal setae. One large, central chloroplast per cell. Measurements: 15-28 um a.a., 22-29 um p.a., 8-10 um ap. (PI. 32, Fig. 1) EM. The chains are linked by fusion of the setae (PI. 32, Figs 2, 3). The valves are weakly silicified, with the flattened, hollow rimoportula coinciding with the slightly excentric annulus (PI. 32, Figs 4-6; PI. 33, Figs 1, 2). The setae are circular in cross-section, with spirals of spines and rows of poroids; terminal and intercalary setae appear to be similar (PI. 33, Figs 3-5). DISTRIBUTION. 17,23. 12. Section DIADEMA (Ehrenb.) Ostenf. emend. Gran One chloroplast. Resting spores with spines on one valve. 3 1 . Chaetoceros diadema (Ehrenb.) Gran Gran, 1905: 84, figs 102a, b; Lebour, 1930: 139, fig. 102; Gran & Angst, 1931: 478, fig. 61; Cleve-Euler, 1951: 103, figs 207a-g; Hendey, 1964: 128, pi. 10, fig. 1; Brunei, 1970: 122, pi. STUDY OF CHAETOCEROS SPECIES 39 Plate 29 Chaetoceros afflnis van willei (Figs 1-3), C. vanheurcki (Figs 4, 5), and C. constrictus (Fig. 6). Fig. 1 : terminal part of chain. LM. Fig. 2: terminal valve. SEM. Fig. 3: detail of terminal seta. SEM. Fig. 4: middle part of chain. LM. Fig. 5: sibling valves showing constrictions at girdle zones and broken setae. SEM. Fig. 6: part of chain. LM. Figs 1,4, 6, bar = 20 urn; Fig. 2, bar = 10 (im; Figs 3, 5, bar = Sum. 40 D. U. HERNANDEZ-BECERRIL Plate 30 Chaetoceros distant. Fig. 1 : complete chain. LM. Fig. 2: part of chain. SEM. Fig. 3: single valve. TEM. Fig. 4: cell showing annulus and costae on valve face. SEM. Fig. 5: two sibling valves with rather wide aperture. SEM. Fig. 6: valve face with central annulus and radiating costae. SEM. Fig. 1 , bar = 50 um; Figs 2, 5, bar = 20 um; Figs 3, 4, 6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 41 Plate 31 Chaetoceros distans. Fig. 1 : view of valve with annulus and costae. TEM. Fig. 2: sibling valves showing wide aperture and fusion of setae. SEM. Fig. 3: Internal view of rimoportula (arrow). SEM. Fig. 4: external view of rimoportula. SEM. Fig. 5: middle part of intercalary seta. SEM. Figs 6, 8: middle part of intercalary seta. Note randomly distributed poroids. TEM. Fig. 7: middle part of intercalary seta. SEM. Figs 1 , 3, bar = 5 um; Fig. 2, bar = 10 urn; Figs 4-8, bar = 2 um. 42 D. U. HERNANDEZ-BECERRIL Plate 32 Chaetoceros brevis. Fig. 1 : typical chain. LM. Fig. 2: short chain. SEM. Fig. 3: part of chain. SEM. Fig. 4: terminal valve showing rimoportula. SEM. Fig. 5: sibling cells showing fusion of setae. SEM. Fig. 6: two sibling valves. SEM. Figs 1 , 2, bar = 20 urn; Fig. 3, bar = 10 urn; Figs 4-6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 43 Plate 33 Chaetoceros brevis. Fig. 1 : terminal valve with excentric rimoportula (arrow). SEM. Fig. 2: terminal valve showing rimoportula. TEM. Fig. 3: detail of seta close to tip. SEM. Fig. 4: middle part of intercalary seta. SEM. Fig. 5: middle part of terminal seta. SEM. Figs 1 , 3-5, bar = 2 |im; Fig. 2, bar = 5 \\m. 27, fig. 4, pi. 30, figs 1-5; Hargraves, 1972: 249, figs 2A-M, 3-14; Drebes, 1974: 77, figs 60, 61; Rines & Hargraves, 1988: 76, figs 150, 151,153. Chaetoceros subsecundus (Grunov) Hust. Hustedt, 1930: 709, fig. 404; Cupp, 1943: 130, fig. 83. Chaetoceros diadema Ehrenb. sensu Meunier, 1913: 33, pi. 5, figs 1-9. Pis 34, 35. LM. Chains straight or slightly twisted, long, delicate. In girdle view, cells rectangular, with sharp corners. Valve face flat or slightly concave, mantle low, lines of girdle straight; apertures of variable width. Setae thin, arising from close to the corners. Terminal setae thicker, directed to the chain axis; intercalary setae diverging at random. One central chloroplast per cell. Resting spores with one valve flattened, the other concave, bearing many spines at the centre. Measurements: 24-42 urn a.a., 20-38 urn p.a., 5-8 urn ap. (PI. 34, Figs 1,2) EM. The valves are lightly silicified (PI. 34, Figs 3, 4), with a central or slightly excentric annulus (PI. 34, Fig. 6) from which weak costae radiate, with no poroids between (PI. 34, Figs 5, 6; PI. 35, Figs 1, 4). Small spines which appear to be solid and branched are sometimes found on the valve face (PI. 34, Figs 3, 5; PI. 35, Figs 3, 5). Also evident is the presence of a hyaline rim on the valve edge and some snail-like projections on the apical regions of the mantle (PI. 35, Fig. 2). The rimoportula is a hollow structure with a very short projection to the outside (PI. 35, Fig. 6). Setae are circular at the base, becoming four-sided distally with spines on the edges. The walls of the setae are regularly perforated by longitudinal rows of poroids; terminal setae do not differ in structure from other setae, but are thicker (PI. 35, Figs 7, 8). No resting spores were examined. DISTRIBUTION. 52,58. REMARKS. All Hargraves' (1972) observations are confirmed here. 32. Chaetoceros seiracanthus Gran Gran, 1897: 21, pi. 3, figs 39^1; Peragallo & Peragallo, 1897-1908: 479, pi. 133, figs 7, 8; Hustedt, 1930: 711, fig. 405; Gran & Angst, 1931: 478, fig. 62; Cleve-Euler, 1951: 104, fig. 209; Hendey, 1964: 129, pi. 15, fig. 1; Rines & Hargraves, 1988: 92, fig. 199. LM. Chains straight, fairly short (6 cells). In girdle view, cells nearly rectangular, mantle low, with constrictions at girdle zones. Valve face flat or concave; apertures relatively wide. 44 D. U. HERNANDEZ-BECERRIL Plate 34 Chaetoceros diadema. Fig. 1 : part of chain. LM. Fig. 2: typical resting spore. LM. Fig. 3: terminal cell of chain. SEM. Fig. 4: another ter- minal cell. SEM. Fig. 5: two sibling valves with costae. TEM. Fig. 6: valve view of valve showing central annulus and costae. TEM. Fig. 1 , bar = 50 Urn; Fig. 2, bar = 20 \an\ Figs 3-6, bar = 10 urn. STUDY OF CHAETOCEROS SPECIES 45 Plate 35 Chaetoceros diadema. Fig. 1 : terminal valve with snail-like structures in apical regions. TEM. Fig. 2: fusion of sibling setae. Note snail-like structure (arrow). SEM. Fig. 3: valve face showing small spines. SEM. Fig. 4: closer view of terminal valve. TEM. Fig. 5: detail of valve face with some spines (arrow). TEM. Fig. 6: rimoportula at terminal valve. SEM. Fig. 7: middle part of seta. TEM. Fig. 8: middle part of seta. SEM. Fig. 1, bar = 10 urn; Figs 2-4, bar = 5 um; Figs 5-8, bar = 2 um. 46 D. U. HERNANDEZ-BECERRIL Sibling cells not touching at the corners, the setae arising from close to the corners and meeting corresponding sibling setae. Setae thin; intercalary setae diverging either randomly or in parallel to the apical axis; terminal setae with similar morphology, directed towards the chain axis. One chloroplast per cell. Measurement: 14-19 um a.a., 12-17 urn p.a., 4-8 urn ap. EM. No observations were made. DISTRIBUTION. 23. characters are very variable (Moreno et al., 1993). Resting spores may provide new evidence for maintaining the two taxa as separate species, but none has yet been reported. On the basis of my observations and taking into account a recent report (Moreno et al., 1993) where no significant differences in structure between the species were seen, C. laevis is considered here as a synonym of C. diversus. Another closely related species is C. rudis Leud.-Fortm., which was poorly described and illustrated. 13. Section DIVERSA Ostenf. Short chains. Intercalary setae of two types, as in section Compressa, but special setae not twisted. One chloroplast. Resting spores unknown. 33. Chaetoceros diversus Cleve Cleve, 1873a: 9, pi. 2, fig. 12; Peragallo & Peragallo, 1897-1908: 487, pi. 135, fig. 4; Hustedt, 1930: 716, fig. 409; Allen & Cupp, 1935: 142, fig. 71; Cupp, 1943: 132, fig. 87; Subrahmanyan, 1946: 142, figs 241-243, non 235; Cleve-Euler, 1951: 101, fig. 197; Hendey, 1964: 130, pi. 17, fig. 4;Navarro, 1982:312,figs26,27;Desikachary&Prema, 1987: pi. 257, figs 1, 2, 4, 6, pi. 262, figs 1,2; Moreno et al., 1993: 420, figs 2-28. Chaetoceros laevis Leud.-Fortm. Leuduger-Fortmorel, 1892: 38, pi. 6, fig. 2. Pis 36-39. LM. Chains straight, short. In girdle view, cells rectangular or nearly square. Valve face slightly concave, mantle high, lines of girdle straight; apertures very narrow. In valve view, cells elliptical. Different kinds of setae present; intercalary setae thin and short; terminal setae thicker, directed to the chain axis. Special intercalary setae curve strongly, thicker and longer than other setae. One large chloroplast present in the centre of the cell. Measurements: 10-22 um a.a., 7-19 um p.a., 2 um ap. (PI. 36, Fig. 1; PL 38, Fig. 1) EM. The chains are short and joined by fusion of the setae (PI. 36, Figs 2, 3; PI. 38, Figs 2, 3). The same general pattern of morphology as shown by most species belonging to subgenus Hyalochaete is present in this species, although there is no clear annulus (PI. 37, Figs 2, 3). The costae become apparent and slightly thicker close to the mantle edge (PI. 37, Fig. 3; PI. 38, Fig. 6). The apertures are sometimes covered by a thin wall and a hyaline rim is found on the edge of some valves (PI. 36, Fig. 4; PI. 38, Figs 4, 6). One small, circular rimoportula is commonly located at the centre of terminal valves; this is short, hollow, and slightly projected to the outside (PI. 37, Fig. 2; PI. 38, Fig. 5). Girdle bands show no costae, but their walls are randomly perforated by small poroids (PI. 38, Figs 4, 6). All setae appear to have essentially the same structure: circular at the base (PI. 36, Figs 5, 6) and polygonal distally, with spines running in rows and the walls perforated by poroids arranged in rows (PI. 37, Figs 4-7; PI. 39, Figs 3-7). Other, bigger poroids are randomly distributed (PI. 37, Figs 4, 7). Special setae are also polygonal, but much thicker (PI. 37, Figs 4, 5, 8; PI. 39, Figs 4, 6, 7). DISTRIBUTION. 1 7, 22, 23, 48, 5 1 . REMARKS. Many authors (e.g. Hustedt, 1930; Cupp, 1943) have discussed the identity of C. laevis and its relationship to C. diversus. The main differences between the two taxa are the direction and structure of the setae, but apparently these 34. Chaetoceros messanensis Castrac. Castracane, 1875: 394, pi. 6, fig. 1; Peragallo & Peragallo, 1897-1908: 488, pi. 129, fig. 1; Hustedt, 1930: 718, fig. 410; Hendey, 1937: 304; Cupp, 1943: 133, figs 89A-B; Cleve-Euler, 1951: 101, figs 198a-c; Hendey, 1964: 129, pi. 12, fig. 3; Okuno, 1970: pi. 654; Evensen & Hasle, 1975: 162, figs 70-74; Desikachary & Prema, 1987: pi. 261, figs 1, 4, 5. Chaetoceros furca Cleve Cleve, 1897: 21, pi. 1, fig. 10. Pis 40, 41. LM. Chains straight, usually short, but sometimes long. In girdle view, cells rectangular, with sharp corners. Valve face flat or slightly concave, mantle low, lines of girdle straight; apertures wide. In valve view, cells elliptical. Setae of three different kinds; terminal setae thin and short, one curved strongly, the other diverging widely; common intercalary setae also thin and slightly diverging; special setae thicker, fusing two sibling setae together for a considerable distance from the chain, then diverging opposite each other, bearing minute spines at the ends. One central chloroplast present. Measurements: 20-32 um a.a., 12-18 um p.a., 5-12 um ap. (PI. 40, Fig. 1) EM. Most of the chains observed were broken (PI. 40, Figs 2, 3). The valves have a pattern of strong, sometimes branched costae which radiates from the thickened centre and may be parallel to the apical axis (PI. 40, Figs 4, 5; PI. 41, Fig. 2). Two or three rows of poroids are evident between the costae. Hyaline rims may be found on the edge of some valves (PI. 41, Figs 1, 2). The rimoportula is excentric and consists of a flattened and elongated tube with a short external projection (PI. 41 , Figs 1 , 3). Terminal and intercalary setae (common type) are circular throughout, with spirals of spines and poroids. Special setae appear to be more heavily silicified, but like the other setae are circular and are somewhat sparsely perforated by randomly spaced poroids (PI. 41, Fig. 7). When the setae finally diverge, they bear spines in spirals, with both relatively large and small poroids scattered at random (PI. 41, Figs 4-6). Their tips appear to be open to the outside. DISTRIBUTION. 40, 42, 44, 45, 48, 65. REMARKS. All previous observations made by Okuno (1970) and Evensen & Hasle (1975) are confirmed here and no additional new findings are reported. The position of this species within section Diversa should, however, be considered carefully because C. messanensis appears to be distinct from C. diversus', the latter develops special setae not only in intercalary valves, but also in terminal valves, and the terminal valve in C. messanensis shows a rather large external projection of the rimoportula and a pattern of costae, lacking in C. diversus. STUDY OF CHAETOCEROS SPECIES 47 Plate 36 Chaetoceros diversus. Fig. 1 : complete chain. LM. Figs 2, 3: complete chains. SEM. Fig. 4: cell in girdle view. Note partially occluded aper- ture. SEM. Fig. 5: chain of two cells. SEM. Fig. 6: detail of fusion of setae. TEM. Fig. 1, bar = 20 urn; Figs 2, 3, bar = 50 urn; Figs 4, 6, bar = 5 urn; Fig. 5, bar = 10 urn. 48 D. U. HERNANDEZ-BECERRIL Plate 37 Chaetoceros diversus. Fig. 1 : sibling valves showing very narrow aperture. TEM. Fig. 2: terminal valve showing rimoportula (arrow). SEM. Fig. 3: valve showing weak costae. TEM. Fig. 4: middle part of special seta. TEM. Fig. 5: middle part of special seta. SEM. Fig. 6: seta close to tip. SEM. Fig. 7: seta close to tip. TEM. Fig. 8: special seta (above) and intercalary seta (below). SEM. Figs 1 , 2, bar = 5 urn; Figs 3-8, bar = 2 jun. STUDY OF CHAETOCEROS SPECIES 49 Plate 38 Chaetoceros diversus. Fig. 1 : typical chain. LM. Figs 2, 3: two chains. SEM. Fig. 4: two sibling valves with aperture covered by thin wall. TEM. Fig. 5: terminal valve with rimoportula. SEM. Fig. 6: detail of valve and girdle bands. TEM. Fig. 1 , bar = 50 urn; Figs 2, 3, bar = 20 urn; Fig. 4, bar = 10 urn; Figs 5, 6, bar = 5 urn. 50 D. U. HERNANDEZ-BECERRIL Plate 39 Chaetoceros diversus. Fig. 1 : part of chain showing rimoportula. SEM. Fig. 2: detail of annulus and rimoportula. SEM. Fig. 3: complete special seta. SEM. Fig. 4: base of special seta. TEM. Fig. 5: view of special seta. SEM. Fig. 6: middle part of special seta. TEM. Fig. 7: special seta close to tip. SEM. Figs 1 , 3, 5, bar = 5 \an; Figs 2, 4, 6, 7, bar = 2 ^m. STUDY OF CHAETOCEROS SPECIES 51 Plate 40 Chaetoceros messanensis. Fig. 1 : view of typical chain. LM. Fig. 2: part of chain. SEM. Fig. 3: part of chain. Note special setae. SEM. Fig. 4: terminal valve. Note direction of terminal setae. TEM. Fig. 5: detail of terminal valve with rimoportula (arrow). TEM. Fig. 6: sibling valves, the lower showing several girdle bands. SEM. Fig. 1, bar = 20 urn; Figs 2, 3, bar = 50 urn; Fig. 4, bar = 10 urn; Figs 5, 6, bar = 5 urn. 52 D. U. HERNANDEZ-BECERRIL Plate 41 Chaetoceros messanensis. Fig. 1 : terminal valve showing flattened, tubular rimoportula. SEM. Fig. 2: sibling valves with costae. TEM. Fig. 3: terminal valve with prominent rimoportula. SEM. Fig. 4: special setae as they diverge. SEM. Fig. 5: detail of special seta. Note spirals of spines. SEM. Fig. 6: tip of special seta (apparently open, or broken?). TEM. Fig. 7: middle part of special seta. The two sibling setae are fused together. TEM. Fig. 1, bar = 10 urn; Figs 2-4, bar = 5 urn; Figs 5-7, bar = 2 urn. STUDY OF CHAETOCEROS SPECIES 14. Section BREVICATENATA Gran Small forms. Chains straight. Terminal setae thinner than other setae. Chloroplasts one or two. 35. Chaetoceros pseudocrinitus Ostenf. Ostenfeld, 1901: 300, fig. 11; Hustedt, 1930: 733, fig. 422; Gran & Angst, 1931: 480, fig. 64; Hendey, 1937: 304; Cleve-Euler, 1951: 1 06, figs 2 17a-c; Hendey, 1964: 132, pi. 14, fig. 2; Rines & Hargraves, 1988: 87, figs 174, 177, 219. LM. Chains straight or somewhat twisted, fairly long. In girdle view, cells rectangular, with sharp corners touching sibling valves. Valve face flat or convex; apertures very narrow. Intercalary setae very thin, arising at random from the valve corners; terminal setae essentially similar, but longer and orientated to the chain axis. One chloroplast per cell. No resting spores seen. Measurements: 10-12uma.a.,9-12ump.a., l-2um ap. EM. No observations were made. DISTRIBUTION. 53. REMARKS. Brunei (1970) has already pointed out the difficulty in identifying the species commonly placed in section Brevicatenata, and the poor present state of our knowledge of them. Brunei (1970) and Rines & Hargraves (1988) discussed the close relationship between C. crinitus F. Schutt, C. ingolfianus Ostenf., and C. pseudocrinitus. Furthermore, Rines & Hargraves (1988) considered C. ingolfianus to be merely a synonym of C. pseudocrinitus. Investigations of their life cycle, as well as observations using EM, are strongly recommended to elucidate this problem. 36. Chaetoceros wighami Brightw. Brightwell, 1856: 108, pi. 7, figs 19-36; Peragallo & Peragallo, 1897-1908: 491, pi. 134, fig. 4; Hustedt, 1930: 724, fig. 414; Cupp, 1943: 136, fig. 91; Cleve-Euler, 1951: 107, figs 221a-c; Hendey, 1964: 131, pi. 11, fig. 3. PI. 43, Fig. 6. LM. Chains straight, short, delicate. In girdle view, cells rectangular, the pervalvar axis being, corners sharp. Valve face concave, mantle low; apertures narrow, elliptical or oblong. In valve view, cells elliptical. Intercalary setae thin, long, emerging from the corners of the valves and diverging following the apical axis; terminal setae curving smoothly at the base, then directed to the chain axis. One chloroplast per cell. Measurements: 6-8 urn a.a., 12-17 urn p.a., 3^ urn ap. (PI. 43, Fig. 6) EM. No observations were made. DISTRIBUTION. 51. REMARKS. Rines & Hargraves (1988) have discussed previous claims about the close relationship between C. wighami and C. amanita A. Cleve, the latter being a fresh- or brackish-water form. 15. Section CURVISETA Ostenf. emend. Gran Chains generally curved. Setae generally directed to the same side. One chloroplast. 53 37. Chaetoceros curvisetus Cleve Cleve, 1889: 55; Peragallo & Peragallo, 1897-1908: 479, pi. 129, figs 4-6; Hustedt, 1930: 737, fig. 428; Cupp, 1943: 137, fig. 93; Subrahmanyan, 1946: 143, figs 238, 244-246; Cleve-Euler, 1951: 100, figs 194a-d; Hendey, 1964: 133, pi. 17, fig. 6; Drebes, 1974: 79, figs 63a, b; Evensen & Hasle, 1975: 159, figs 23-26; Navarro, 1982: 309, figs 18, 19; Desikachary et al., 1987: pi. 502, fig. 7; Rines & Hargraves, 1988: 71, figs 141, 142 (non Chaetoceros curvisetus Cleve sensu Hustedt, 1920: pi. 327, figs 9, 10 = C. pseudocurvisetus L. Mangin). Chaetoceros secundus Cleve sensu Gran & Angst, 193 1 : 48 1 , fig. 66. Pis 42, 43, Figs 1-5. LM. Chains curved, usually long. In girdle view, cells rectangular, with sharp corners. Valve face concave, mantle low, lines of girdle straight; apertures wide. In valve view, cells nearly circular to elliptical, the setae curving to the same direction. Intercalary setae rather thin, arising from the valve corners and diverging in the apical plane transapically; terminal setae slightly shorter, directed to the chain axis. One central chloroplast present. Resting spores spineless, with one valve flat, the other convex. Measurements: 15-22 urn a.a., 19-25 um p.a., 12-17 um ap. (PI. 42, Figs 1,2) EM. The chains are linked by fusion of the setae (PI. 42, Figs 3, 4). The valve pattern follows the general morphology for Hyalochaete species (PI. 42, Figs 4-6; PI. 43, Fig. 2), but in this species there is also a hyaline rim on the valve edge (PI. 42, Fig. 5). The annulus may be central or excentric (PI. 43, Figs 1, 2). The rimoportula is a flattened, hollow structure, located near the centre of the annulus (PI. 43, Fig. 3). The setae bear a spiral pattern of spines and rows of poroids in the wall, as well as some scattered, larger poroids (PI. 43, Figs 4, 5). Although terminal setae have the same morphology as intercalary setae, they are slightly twisted at the base. DISTRIBUTION. 23, 42, 48, 50, 64. REMARKS. All previous observations (Evensen & Hasle, 1975) on this species are confirmed here. 38. Chaetoceros pseudocurvisetus L. Mangin Mangin, 1910: 350, figs 311, 411; Gran & Yendo, 1914: 22, figs 12a, b; Hustedt, 1930: 739, fig. 427; Allen & Cupp, 1935: 142, fig. 73; Cupp, 1943: 138, fig. 94; Hendey, 1964: 134, pi. 18, fig. 1; Fryxell, 1978: 68, figs 22-26; Navarro, 1982: 316, figs 37-39. Pis 44, 45. LM. Chains curved, short or long. In girdle view, cells rectangular, with smooth corners and processes near the corners connecting sibling cells in the chain. Valve face concave, mantle low, lines of girdle straight or slightly concave; main apertures narrow, elliptical, with two small apertures present at corners beside the main aperture. In valve view, cells nearly circular to elliptical, the setae parallel and curving in the same direction or curving in opposite directions. Setae generally thin, arising from the valve corners and curving transapically (girdle view); terminal setae differ in their direction, diverging to the chain axis. One large, central chloroplast present. Measurements: 13-21 um a.a., 16-22 um p.a., 4-6 um ap. (PI. 44, Figs 1, 2) EM. Partially broken chains were observed (PI. 44, Figs 3, 5). The valves are not heavily silicified and have the same general pattern described earlier (e.g. for C. decipiens). However, in this 54 D. U. HERNANDEZ-BECERRIL Plate 42 Chaetoceros curvisetus. Fig. 1 : chain in narrow girdle view. LM. Fig. 2: chain in broad girdle view. LM. Fig. 3: part of chain. SEM. Fig. 4: detail of cells. SEM. Fig. 5: sibling cells showing aperture, hyaline rim on valve edge, and girdle bands. SEM. Fig. 6: two sibling valves. TEM. Figs 1-3, bar = 20 urn; Fig. 4, bar = 10 urn; Figs 5, 6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 55 ife iH Plate 43 Chaetoceros curvisetus (Figs 1-5), C. wighami (Fig. 6), and C. gracilis (Figs 7, 8). Fig. 1 : sibling valves with central annulus. SEM. Fig. 2: valve view of valves with central annulus and radiating costae. TEM. Fig. 3: terminal valve showing rimoportula (arrow). SEM. Fig. 4: middle part of seta. Note spirals of spines. SEM. Fig. 5: middle part of seta showing spirals of small poroids and larger, scattered poroids. TEM. Fig. 6: typical chain. LM. Fig. 7: collapsed cell showing setae and rimoportulae (one on each valve). SEM. Fig. 8: detail of seta. SEM. Figs 1-3, bar = 5 urn; Figs 4,5,8, bar = 2 urn; Fig. 6, bar = 20 urn; Fig. 7, bar = 1 0 um. 56 D. U. HERNANDEZ-BECERRIL Plate 44 Chaetocerospseudocurvisetus. Fig. 1 : typical chain in narrow girdle view. LM. Fig. 2: chain in broad girdle view. LM. Fig. 3: part of chain. SEM. Fig. 4: valve showing annulus and costae. TEM. Fig. 5: part of chain showing small apertures in apical areas. SEM. Fig. 6: two sibling valves. SEM. Figs 1-3, bar = 20 \m\; Figs 4-6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 57 Plate 45 Chaetoceros pseudocurvisetus. Fig. 1 : sibling valves showing annulus and small apertures at apices. SEM. Fig. 2: detail of apical areas from inside. SEM. Fig. 3: terminal cell with excentric rimoportula (arrow). SEM. Fig. 4: detail of rimoportula. Note small raised area in centre of annu- lus. SEM. Fig. 5: another view of rimoportula. SEM. Fig. 6: middle part of seta. TEM. Fig. 7: tip of seta. TEM. Fig. 8: tip of seta. Note spirals of spines. SEM. Figs 1 , 3, bar = 5 urn; Figs 2, 4-8, bar = 2 urn. 58 species no poroids could be observed (PI. 44, Fig. 4). At the edge of the valve, close to where the setae arise, there are four small apertures which are formed by four processes that connect sibling cells (PI. 44, Fig. 6). Inside there are two depressions, each without costae or thickenings on the apex (PI. 44, Fig. 6; PI. 45, Figs 1, 2). The setae arise between. The rimoportula coincides with the annulus which is located slightly to one side of the valve (PI. 45, Fig. 3). The rimoportula consists of an elongate hollow which protrudes to the outside. Close to the rimoportula there is a small, round, raised area at the very centre of the annulus (PI. 45, Figs 4, 5). The setae show the same pattern as C. curvisetus, being circular throughout (PI. 45, Figs 6-8). The tips are sharply pointed. DISTRIBUTION. 45,64. REMARKS. The morphology of this species, as described here, is consistent with Fryxell's (1978) observations, although new information is presented for the rimoportula. 39. Chaetoceros debilis Cleve Cleve, 1894: 13, pi. 1, fig. 2; Meunier, 1913: 43, pi. 7, figs 1-11; Hustedt, 1930: 740, fig. 428; Gran & Angst, 193 1:481, fig. 65; Hendey, 1937: 305; Cupp, 1943: 138, fig. 95; Hendey, 1964: 133, pi. 14, fig. 7; Brunei, 1970: 130, pi. 32, figs 1-6; Drebes, 1974: 81, figs 64a, b; Evensen & Hasle, 1975: 159, figs 27-32; Rines & Margraves, 1988: 72, figs 143, 147. Pis 46, 47. LM. Chains curved and twisted, generally long. In girdle view, cells rectangular, with sharp corners, the apical axis usually longer. Valve face flat or concave, mantle low, lines of girdle straight; apertures rather wide. In valve view, cells elliptical, with all setae curving smoothly in the same direction (in the apical plane, transapically). Intercalary setae thin; terminal setae shorter and slightly thicker. One central chloroplast present. Measurements: 17-28 um a. a., 5-15 urn p.a., 5-8 um ap. (PI. 46, Figs 1,2) EM. The chains are curved and slightly twisted (PI. 47, Figs 3, 4). The valves appear to be thinly silicified, with the same basic morphology as other species of subgenus Hyalochaete. The annulus is commonly excentric (PI. 46, Fig. 5). The costae may be branched at times (PI. 47, Fig. 4) and hyaline rims are found on the edges of some valves (PI. 47, Figs 1, 2). The rimoportula resembles that of C. curvisetus, but is perhaps more excentrically positioned (PI. 47, Figs 1-5). The setae are circular in cross-section and follow the pattern of C. curvisetus. Terminal setae are thicker and shorter than other setae, polygonal at the base and with no spines (PI. 47, Figs 6-8). The setae tips are flattened. DISTRIBUTION. 23, 24, 28, 52. REMARKS. The three species studied here from section Curviseta are closely related. All observations made here agree with those made previously (e.g. Evensen & Hasle, 1975). 16. Section ANASTOMOSANTIA Ostenf. Chains mostly loose. Setae united by a bridge. 40. Chaetoceros rectus Hern. -Bee. Hernandez- Becerril, 19926: 219, figs 2, 16-29. DISTRIBUTION. 23,24. D. U. HERNANDEZ-BECERRIL 17. Section FURCELLATA Ostenf. Chains generally loose, without differentiated terminal setae. One chloroplast. 4 1 . Chaetoceros radicans F. Schiitt Schutt, 1895: 48, pi. 5, fig. 27; Hustedt, 1930: 746, fig. 431; Gran & Angst, 1931: 483, fig. 67; Cupp, 1943: 141, fig. 97; Brunei, 1970: 132, pi. 33, figs 1-4, pi. 34, figs 1, 2; Hendey, 1964: 134, pi. 14, fig. 4; Drebes, 1974: 82, fig. 66; Fryxell & Medlin, 1981: 8, figs 9-15, 29^12; Rines & Hargraves, 1988: 90, figs 192-1 98. Pis 48, 49. LM. Chains straight, curved or slightly twisted, generally long, delicate. In girdle view, cells rectangular to oblong, with rounded corners, the sibling cells not touching. Valve face convex (dome-shaped), mantle low to high, lines of girdle straight; apertures narrow. In valve view, cells elliptical, with setae diverging in the transapical axis. Setae thin, bearing numerous small spines; they arise from the corners of the valves and diverge nearly perpendicular to the chain axis. Intercalary and terminal setae undifferentiated. One large chloroplast in the centre of the cell. Measurements: 13-24 um a. a., 10-22 um p.a., 3-5 um ap. (PI. 48, Fig. 1) EM. The chains are delicate and slightly twisted (PI. 48, Fig. 2). The valves are weakly silicified (PI. 48, Fig. 3), having a quite apparent, centrally placed annulus (PI. 48, Fig. 4). Some costae converge at the apical regions, making the valve slightly thicker (PI. 49, Figs 2, 3). No poroids were seen, or they were too small to be detected. A hyaline rim is present on the edge of most of the valves (PI. 48, Fig. 6; PI. 49, Fig. 4). The rimoportula is slightly excentric and consists of a flattened and elongated hollow orientated in the apical axis or diagonal, being a tubular structure to the outside (PI. 49, Figs 2-5). The setae are orientated to the transapical axis in opposition to each other, especially on terminal valves (PI. 49, Figs 1 , 2, 5). The setae seem to be circular throughout, developing spines which are sometimes branched, usually after fusion with adjacent setae (PI. 48, Fig. 5); these spines appear to be solid (PI. 49, Fig. 6). It is not possible to distinguish any particular pattern of spine distribution, so they are probably positioned at random (PI. 48, Figs 3, 4, 6; PI. 49, Fig. 7). The setae walls are simply perforated by minute poroids and appear thicker than those of the valves (PI. 49, Fig. 6). DISTRIBUTION. 7, 1 1, 17, 18, 23, 28, 40-42, 44, 45, 48, 52. REMARKS. All Fryxell & Medlin's (1981) observations on this species agree with those presented here; however, no details of the resting spores are reported in this study. The rimoportula of the specimens given as C. radicans by Helmcke & Krieger (1954: pi. 125) is completely different from that described here, so it is possible their figure does not correspond with this species. 42. Chaetoceros filiferus G. Karst. Hernandez-Becerril, 19936: 170, figs 15-28. DISTRIBUTION. 42, 45, 48. 18. Section SOCIALIA Ostenf. Cells in irregular chains. One chloroplast. Resting spores variable. STUDY OF CHAETOCEROS SPECIES 59 Plate 46 Chaetoceros debilis. Fig. 1 : typical chain. LM. Fig. 2: chain in broad girdle view. LM. Fig. 3: two chains. SEM. Fig. 4: another chain. SEM. Fig. 5: valve showing slightly excentric annulus and costae. TEM. Fig. 6: detail of fusion of setae. SEM. Figs 1, 3, 4, bar = 50 urn; Fig. 2, bar = 20 Urn; Fig. 5, bar = 5 nm; Fig. 6, bar = 2 urn. 60 D. U. HERNANDEZ-BECERRIL Plate 47 Chaetoceros debilis. Fig. 1 : terminal cell with excentric rimoportula (arrow). SEM. Fig. 2: terminal cell showing rimoportula (arrow). Note short, thick terminal setae. SEM. Fig. 3: partially broken terminal valve. Arrow indicates rimoportula. TEM. Fig. 4: detail of rimoportula. SEM. Fig. 5: terminal valve with elongate rimoportula (arrow). SEM. Fig. 6: seta close to tip. SEM. Fig. 7: middle part of seta. TEM. Fig. 8: middle part of seta. SEM. Figs 1-3, 5, bar = 5 um; Figs 4, 6-8, bar = 2 urn. STUDY OF CHAETOCEROS SPECIES 61 Plate 48 Chaetoceros radicans. Fig. 1 : typical chain. LM. Fig. 2: part of chain. SEM. Fig. 3: complete cell in valve view. TEM. Fig. 4: detail of valve. TEM. Fig. 5: internal view of valve. SEM. Fig. 6: terminal valve. SEM. Figs 1 , 2, bar = 50 urn; Fig. 3, bar = 20 nm; Fig. 4, bar = 1 0 jam; Figs 5, 6, bar = 5 nm. 62 D. U. HERNANDEZ-BECERRIL Plate 49 Chaetoceros radicans. Fig. 1 : terminal valve in valve view. SEM. Fig. 2: another terminal valve. Arrow indicates rimoportula. TEM. Fig. 3: detail of valve showing rimoportula. TEM. Fig. 4: rimoportula at terminal valve. SEM. Fig. 5: flattened, hollow rimoportula (arrow). SEM. Fig. 6: detail of seta close to base. TEM. Fig. 7: middle part of seta. SEM. Fig. 1 , bar = 10 urn; Figs 2-5, bar = 5 um; Fig. 6, bar = 2 um. STUDY OF CHAETOCEROS SPECIES 63 43. Chaetoceros socialis Lauder var. ? Lauder, 1864: 77, pi. 8, fig. 1; Meunier, 1913: 46, pi. 7, figs 26-29; Gran & Angst, 1931: 486, fig. 70; Cupp, 1943: 143, fig. 100; Subrahmanyan, 1946: 143, figs 256, 251?; Cleve-Euler, 1951: 102, figs 199a-e; Brunei, 1970: 135, pi. 29, figs 3, 4, pi. 36, figs 1-3, pi. 37, figs 1-3; Evensen & Hasle, 1975: 160, figs 33-39. Pis 50, 51. LM. Cells gathered in short chains or clumps, joined through long setae. In girdle view, cells rectangular, with smooth corners which do not touch sibling cell corners. Valve face concave, mantle low, lines of girdle straight; when present, apertures narrow to wide. In valve view, cells elliptical, the setae curving smoothly, sometimes one longer and straight, following the apical axis. All setae thin, bearing small spines. One chloroplast present at the centre of the cell. Measurements: 10-18 um a.a., 8-12 um p.a., 6-7 um ap. (PI. 50, Fig. 1) EM. Chains or clumps seem to be joined by special long setae (PI. 50, Fig. 2). The valves are thinly silicified (PI. 50, Figs 3-6). The previous characteristic of an annulus with a costae pattern radiating from it, is also present in this species (PI. 51, Fig. 1). Rows of fine poroids are found between the costae (PI. 51, Fig. 1). No rimoportulae were seen; however, it is assumed that one is present just in terminal valves, because in intercalary valves the rimoportula was absent. The setae arise from close to the valve corners and following a short base meet and fuse with sibling setae (PI. 50, Fig. 6; PI. 51, Figs 2, 3). They are circular throughout, generally developing spines after fusion (PI. 50, Fig. 5; PI. 51, Fig. 3); the spines run in spirals and the walls are perforated by spiral rows of fine poroids, as well as some scattered, bigger poroids (PI. 51, Fig. 7, 8). One seta is longer than the others; it is spineless, with longitudinal rows of poroids and also randomly scattered bigger poroids (PI. 51, Fig. 6). Resting spores are generally paired, connected by spineless setae (PI. 51, Figs 4, 5). One valve is strongly convex with processes radiating from the centre; the other valve is slightly concave and smooth (PI. 51, Figs 4, 5). DISTRIBUTION. 42, 52. REMARKS. Margraves (1979) mentioned the possibility that more than one taxa are involved in the species currently known as C. socialis, mainly due to the range of resting spores found and also the arrangement of the chains. The taxon identified here indeed belongs to C. socialis, but it is probably a variety of it. On the other hand, Evensen & Hasle (1975) discussed the difference in structure between the long seta and the others: the long seta lacks spines and the pattern of poroids is straight, distinct from the 'normal' setae. This seta would be involved in the chain formation mechanism 'by sticking together near the tip with setae from other chains or cells' (Evensen & Hasle, 1975). 19. Section SIMPLICIA Ostenf. Cells generally solitary, small and fragile. LM. Cells solitary, small and fragile. In girdle view, cells rectangular, with concave valve face. No clear distinction at girdle zones. In valve view, cells elliptical. Setae very thin, long and delicate, orientated towards the apical axis or smoothly curved. Two chloroplasts are placed opposite the sides of the cell, at girdle zones. Measurements: 7-9 um a.a., 1 1-12 um p.a. EM. The valves are delicate, very thinly silicified (PI. 43, Fig. 7). There was no apparent annulus or costae pattern (perhaps not resolved at the magnifications used); however, a minute external protrusion of the rimoportula was found at the centre of the valve face (PI. 43, Fig. 7). The edges of the valve face show a rim (PI. 43, Fig. 7). Setae borne on the apices of the valves appear to be circular at the base, becoming polygonal away from the valve, with minute spines (in spirals?) (PI. 43, Fig. 8). DISTRIBUTION. 23,50,51. REMARKS. Rines & Hargraves (1988) have thoroughly discussed the problems arising when dealing with species of section Simplicia (e.g. they may or may not be single cells, the descriptions are incomplete, etc.). These authors recommend that C. gracilis is treated 'with extreme caution'. 45. Chaetoceros vistulae Apstein Apstein, 1909: 136, fig. 2; Ostenfeld, 1912: 10, fig. 24; Hustedt, 1930: 762, fig. 443; Cupp, 1943: 144, fig. 102. PI. 52. LM. Cells solitary, fragile. In girdle view, cells rectangular, with sharp corners, the pervalvar axis being longer. Valve face flat or slightly concave, mantle low, lines of girdle straight. In valve view, cells elliptical. Setae thin, long, arising from corners of the valves, directed towards the pervalvar axis, at an angle of about 10-20° from this axis. One large chloroplast present in each cell, placed centrally. Measurements: 6-9 urn a.a., 20-27 um p.a. (PI. 52, Fig. 1) EM. Valves extremely weakly silicified (PI. 52, Figs 2, 3), with annulus and costae well apparent (PI. 52, Fig. 4). Some spine-like projections arise from the centre of the valve and a hyaline rim appears on the valve edge (PI. 52, Figs 4, 6). There is one rimoportula per valve, each centrally located and consisting of a flattened tube open to the outside (PI. 52, Fig. 6). The setae are circular at the very base (PI. 52, Fig. 4), becoming polygonal (four or five-sided) distally, with rows of minute spines running along each edge and a single longitudinal row of poroids running along each side (PI. 52, Fig. 5). DISTRIBUTION. 16, 23, 24, 51. REMARKS. Details from electron microscopy of the valve (costae) and setae of a closely related species, C. gracilis, described above and also by Rogerson et al. (1986), reveal similar morphology to C. vistulae as shown here. Apparently, this is a brackish-water species, but it has also been recorded from marine waters (e.g. Cupp, 1943). 44. Chaetoceros gracilis F. Schiitt Schiitt, 1895: 42, fig. 13; Hustedt, 1930: 758, fig. 440; Gran & Angst, 1931: 487, fig. 71; Cupp, 1943: 143, fig. 101; Cleve-Euler, 1951: 108, figs 225a-e; Hendey, 1964: 137, pi. 14, fig. 6. PI. 43, Figs 7, 8. III. Subgenus BACTERIASTROIDEA Hern.-Bec. Valves cylindrical. Each valve possessing three pairs of setae, two pairs very reduced, the other pair usually thick. Regular projections (outgrowths) present on the valve edge. Chloroplasts two. 64 D. U. HERNANDEZ-BECERRIL Plate 50 Chaetoceros socialis var. ? Fig. 1 : complete chain. LM. Fig. 2: part of chain. SEM. Fig. 3: sibling valves. TEM. Fig. 4: sibling valves. Note direction of setae. SEM. Fig. 5: sibling valves. Note direction of setae. TEM. Fig. 6: internal view of valve. SEM. Figs 1,2, bar = 50 nm; Figs 3, 4, bar = 20 nm; Fig. 5, bar = 10 nm; Fig. 6, bar = 5 urn. STUDY OF CHAETOCEROS SPECIES 65 Plate 51 Chaetoceros socialis var. ? Fig. 1 : detail of valve showing annulus and costae. TEM. Fig. 2: internal view of valve. SEM. Fig. 3: two sibling valves. Note fusion of setae. SEM. Fig. 4: resting spores in pairs. SEM. Fig. 5: girdle view of resting spores. Note processes in upper valve. SEM. Fig. 6: various long setae. SEM. Fig. 7: middle part of seta. TEM. Fig. 8: middle part of seta. SEM. Figs 1, 2, 4, 5, bar = 5 urn; Figs 3, 6, bar = 10 um; Figs 7, 8, bar = 2 um. 66 D. U. HERNANDEZ-BECERRIL Plate 52 Chaetoceros vistulae. Fig. 1 : typical cell (solitary). LM. Figs 2, 3: two complete cells. SEM. Fig. 4: detail of valve face. Arrow indicates rimoportula. SEM. Fig. 5: middle part of seta. SEM. Fig. 6: detail of prominent rimoportula and small scattered spines. SEM. Figs 1,3, bar = 20 Urn; Fig. 2, bar = 10 (im; Fig. 4, bar = 5 urn; Figs 5, 6, bar = STUDY OF CHAETOCEROS SPECIES 67 46. Chaetoceros bacteriastroides G. Karst. Hernandez-Becerril, 1993a: 119, figs 1-15. DISTRIBUTION. 15,24. REMARKS. The structure of this species, as already mentioned, differs from the general morphology of species commonly included in the subgenera Chaetoceros (Phaeoceros) and Hyalochaete. This has been taken into account to propose the subgenus Bacteriastroidea, where C. bacteriastroides remains isolated due to a lack of closely related species. DISCUSSION Morphology General accounts of the morphology of Chaetoceros species have been published by several authors, the most recent overview of the genus being provided by Evensen & Hasle (1975). As summarized by Desikachary (1956), the various studies of Chaetoceros using TEM have revealed the structure of the cell wall, which is simply perforated by small poroids with some 'vein-like thickenings' in thin-walled species. These findings were complemented by Okuno's (1956) studies, who added observations on the annulus ('central area'), as well as the costae ('vein-like thickenings') radiating from this field (structures usually found in many species of subgenus Hyalochaete). Evensen & Hasle (1975) discussed the difference in structure between the cell wall in the main part of the valves and that in the setae, with special emphasis on the morphology of setae and rimoportulae. All these characteristics are confirmed here and there appears to be little new information left to be added to the knowledge of the general morphology of Chaetoceros. Many questions have arisen concerning the function of certain structures in Chaetoceros. Evensen & Hasle (1975) emphasized the enigma of the function of the rimoportulae, which at the moment remains a matter for speculation. Ideas put forward for their function in other centric diatoms include differential permeability to certain chemicals (trace elements) (Fryxell & Hasle, 1974), secretion of mucilage pads or stipes for attachment (Crawford, 1975), and role in movement (Andersen et al., 1986; Medlin et al., 1986). Of these hypotheses, only the first might be a possible explanation for their function in Chaetoceros. The presence of multiple rimoportulae in a single valve has recently been found in several species, including Chaetoceros compressus var. hirtisetus Rines & Hargraves (Rines & Hargraves, 1990), C. coarctatus (Hernandez-Becerril, 199 la), C. bermejensis (Hernandez-Becerril, 1991c), and C. buceros G. Karst. (Hernandez-Becerril et al., 1993), and this has been related to the evolution of the raphe (Rines & Hargraves, pers. comm.). The function of the setae has generated more discussion than that of the rimoportulae, as the setae are visible with light microscopy and have been observed since the original type species description was made (Ehrenberg, 1844). Many planktologists have found very elongated or chain-forming species and have believed that in Chaetoceros the setae might have the function of conferring buoyancy. This hypothesis seems very likely for some species, but for attached forms, such as C. dadayi Pavill. and C. tetrastichon, this is clearly not the case (Hernandez-Becerril, 1992a). Many species have tiny, delicate setae and it is difficult to see how these could aid buoyancy. Others have very robust setae which must also be heavy and increase the density of the cell. However, very often these setae contain chloroplasts and may therefore increase the photosynthetic capacity of the cell. There is not enough evidence regarding the distribution of robust forms with chloroplasts in the setae (e.g. subgenus Chaetoceros) as deep water forms, or more fragile, non chloroplast-bearing forms (e.g. subgenus Hyalochaete) as more shallow water forms. The structure of the setae in species belonging to subgenus Hyalochaete, which lack chloroplasts in the setae, reveals a wide range of adaptations for buoyancy; Fryxell (1978) considers that the spiral pattern may be advantageous in relation to buoyancy. The setae of C. decipiens and C. lorenzianus have perforations along their length which appear to be uncovered even when the organisms are living, and are larger than those in the other species of the subgenus. C. socialis var. ?, as described here and previously (Evensen & Hasle, 1975), has one very long, spineless seta which may connect with similar setae to form the chain, although fusion between cells or their parts is rare (Round & Crawford, 1981), but still possible in 'inseparable' chains (Stosch, 1977). The method of chain formation in Chaetoceros has also been discussed (e.g. Rines & Hargraves, 1988). The main ways were outlined by Fryxell (1978) and Fryxell & Medlin (1981) as, for example: fusion of setae (the most typical), fusion of edge valves and setae, holding of setae, and possession of 'prehensors'. More recently, Hernandez-Becerril (19926) mentioned yet another special manner of chain formation, which was found in C. rectus, a new species described therein. A thin and simple perforated, siliceous wall connects the apices of the sibling valves, so the sibling setae pass through this wall without fusion. Rines & Hargraves (1988) commented on the particular mechanism of chain formation in C. rostratus: by fusion of linking spines on the valve face. The valve formation of this species was studied by Li & Volcani (1985), who found that these intercellular linkages are formed by incomplete cytokinesis during cell division, after which silicification occurs and a siliceous septum divides the two adjacent cells. The different mechanisms of chain formation should be considered as one of the most important points for understanding the great diversity of species in Chaetoceros, and one of the main factors in the evolution of the genus. They should also be considered as a valuable taxonomic character, especially when proposals for future classifications are being considered, and in particular those at the sectional level. Solitary forms are encountered in subgenus Chaetoceros (e.g. C. peruvianus, C. pendulus) and subgenus Hyalochaete (e.g. C. gracilis, C. vistulae). Despite the well-apparent differences, it is possible to make a case for parallel evolution having occurred, with these forms having evolved from a chain-forming ancestor. With regard to resting spores, it is clear that a number of well-known species can be recognized using this character, but many species (especially those belonging to subgenus Chaetoceros) lack resting spores. Also, morphological variation does exist in these structures. The study of the resting spores (including mechanisms of production) will certainly provide evidence to increase our understanding of the biology, ecology, and taxonomy of Chaetoceros. Another aspect, which has not been extensively studied, is intraspecific morphological variation. This is considerable in some taxa and may lead to taxonomic confusion. C. diversus has been shown to exhibit a high level of variation, currently involving at least two species (C. diversus - C. laevis), and 68 D. U. HERNANDEZ-BECERRIL perhaps a third (C. rudis) (Moreno et al., 1993). Other species are morphologically very similar to closely related taxa and this may cause identification problems. Therefore, further studies on morphological variation and life cycles are highly recommended to elucidate 'valid' or 'currently recognized species'. The observations made here are summarized in Tables 2 and 3, and general conclusions on the morphology of Chaetoceros species can be made from them. Taxa belonging to subgenus Chaetoceros are robust forms with large and complex appendages, heavily silicified; valves perforated by poroids, with thickenings or costae in some cases; rimoportula in each valve, central or to one side of the valve, tubular, circular, oval or slit-like; setae polygonal, well-armoured with spines, the wall areolated by rows of poroids or striae and costae pattern. Taxa belonging to subgenus Hyalochaete are less robust and many are delicate and fragile, with delicate and tiny appendages; valves and setae are not thickly silicified, the valve face having an annulus from which a pattern of weak or strong costae radiates and rows of poroids run in between the costae; the rimoportula is present at least on terminal valves, being central or excentric, circular, slit-like or a flattened hollow with a true labiate structure inside; setae generally circular, in a few cases polygonal, with short spines running straight or in a spiral, the wall perforated by rows of poroids, also running straight or in a spiral. The monotypic subgenus Bacteriastroidea is characterized by the presence of three pairs of setae per valve, two of these pairs reduced. A combination of certain other characteristics shown by Chaetoceros and Hyalochaete species is also present. Taxonomic relationships As already stated (p. 2), the genus Chaetoceros is included in the family Chaetocerotaceae, together with Bacteriastrum and Gonioceros (Round et al., 1990), although recent studies (Crawford et al., 1994) have left the family with only two members: Chaetoceros and Bacteriastrum. Evensen & Hasle (1975) mentioned the difficulty of relating Chaetoceraceae (including Acanthoceros, following Simonsen, 1979) to other families, and accordingly that it would be left in 'an isolated taxonomic position'. However, taking into account the two families recently established by Round & Crawford (Round et al., 1990): Acanthocerataceae and Attheyaceae, the family Chaetocerotaceae is more easily related, at least with two genera (Acanthoceros and Attheyd) and their corresponding families. Recent studies made on species of the genus Attheya include proposals to transfer two Gonioceros species, G. armatus (West) Peragallo and G. septentrionalis (0strup) R.M. Crawford, to that genus. The new subclass, Chaetocerotophycidae, includes the new order Leptocylindrales, with only one family, Leptocylindraceae M. Lebour, 'tentatively placed until further studies can be made' (Round et al., 1990). This inclusion is not convincing to me, as the genus Leptocylindrus Cleve is completely different to Chaetoceros or Bacteriastrum, although superficially similar (in the girdle bands) to Acanthoceros. A new genus closely related to Chaetoceros was recently described, namely Miraltia D. Marino, Montresor & Zingone (Marino et al., 1987), but it has now been incorporated into Chaetoceros (Marino et al., 1991). Opinions differ on the phylogeny in the family. One view is that the evolutionary trend is from bilateral to radial symmetry (Simonsen, 1979), while another is that the sequence is the opposite (Fryxell, 1978; Fryxell et al., 1986). However, Bacteriastrum remains the most closely related genus to Chaetoceros, and C. bacteriastroides may be an important link in this relationship (see Fryxell, 1978; Hernandez-Becerril, 1993a). Classification The current taxonomic classification has been given above: three subgenera and 19 sections (see p. 2). Rines & Hargraves (1988) recently drew attention to some of the inconsistencies in the traditional classification and the need for a 'contemporary revision'. I fully support this view and also accept that this revision would take some time. However, the present division into three subgenera still seems reasonable and is supported by the various characters mentioned in the summary (p. 2) and by Table 2 The major characteristics of Chaetoceros taxa (subgenera Chaetoceros and Bacteriastroidea) observed by electron microscopy. Species Valve face and mantle Rimoportula features Seta structure: cross-section, wall pattern Remarks on setae C. atlanticus var. neapolitanus 2,4 One, tubular, central 5,7, 2S/2C C. atlanticus var. skeleton 2,4 One, tubular, central 5,8, 2S/2C - C. dichaeta 1,4 One, tubular, central 6,8, poroids - C. octagonus 2 One, circular, central 5,8, 1S/2C - C. densus 3,4 One, circular, central 5,7, 1S/2C - C. tetrastichon 3 One, slit-like, central 5,7, 3S/2C - C. seychellarus 1,4 One, oval, excentric 5,7, 1S/2C T setae fused C. denticulatus 1 One, oval, central 5,8?, 1S/2C - C. rostratus 3 One, circular, excentric 5,7, 2S/2C - C. danicus 2,4 One, oval, central 5,7, 1S/2C - C. concavicornis 1? One, spine-like?, excentric 5,7,? - C. convolutus 1 One, spine-like, excentric 5,7,? - C. peruvianus 1 One, spine-like, excentric 5,7, 3S/2C Ant. T setae fused C. pendulus 3 One, oval, excentric 5,7, 3S/2C - C. coarctatus 1 Numerous, small, central 6,7, 1S/2C T setae thicker C. bacteriastroides 2,4 One, slit-like, central 6,8, poroids XT setae reduced 1 = Valve perforated, no costae, 2 = Valve perforated, weak costae, 3 = Valve perforated, costae, 4 = Thickening (granules), 5 = Setae square (four-sided) in cross-section, 6 = Setae polygonal (more than four sides) in cross- section, 7 = Big spines along edges, 8 = Small spines along edges, S = Striae, C = Costae, T = Terminal, XT = Extra. Taxa of subgenus Chaetoceros bear at least one rimoportula on each valve. STUDY OF CHAETOCEROS SPECIES Table 3 The major characteristics of Chaetoceros taxa (subgenus Hyalochaete) observed by electron microscopy. 69 Species Valve face and mantle Rimoportula features Seta structure: cross-section, wall pattern Remarks on setae C. decipiens 2,4 One, hollow, central 7,9, big poroids T setae thicker C. decipiens f. singularis 2,4 ? 7,9, big poroids - C. lorenzianus 1.4 One, tubular, central 7,9, big poroids - C. lorenzianus f. forceps 1,4? ? 7,9 - C. bermejensis 2,4 Numerous, small, central 6,8, poroids - C. compressus 1,4 One, hollow, excentric 6,8, poroids SP setae thicker C. didymus 1,5 One, hollow, central 7,9, poroids T setae thicker C. protuberans 1,5 One, hollow, central 7,9, poroids T setae thicker C. constrictus - - - - C. vanheurcki ? 9 6,8? - C. affmis var. qffinis 2,4 One, tubular, central 6,8, poroids T setae thicker C. qffinis var. willei - One, excentric 6?,8, poroids? - C. paradoxus 1,5? One, hollow, central 7,9, poroids - C. distans 1,4 One, slit-like, central 7,9, poroids - C. brevis 1,4 One, hollow, excentric 6,8, poroids - C. diadema 1,5? One, hollow, central 7,9, poroids - C. seiracanthus - - - - C. diversus 1,4 One, circular, central 6,9, poroids SP setae thicker C. messanensis 3,4 One, hollow, excentric 6,8, poroids SP setae fused C. pseudocrinitus - - - - C. wighami - - - - C. curvisetus 1,4 One, hollow, excentric 6,8, poroids - C. pseudocurvisetus 1 One, hollow, excentric 6,8, poroids - C. debilis 1,4 One, hollow, excentric 6,8, poroids T setae reduced C. rectus 2,4 One, beak-like, central 6,8, poroids - C. radicans 1 One, hollow, excentric 6, poroids - C.filiferus 2,4 One, elongate, central 6,8, poroids - C. socialis var. ? 1,4 7 6,8, poroids One seta very long C. gracilis ? One, hollow, central 7,9?, poroids? - C. vistulae 1,4 One, tubular, central 7,9, poroids — 1 = Costae pattern, 2 = Weak costae pattern, 3 = Thick costae pattern, 4 = Poroids between costae, 5 = Poroids at peripheral areas, 6 = Setae circular in cross-section, 7 = Setae polygonal in cross-section, 8 = Spiral pattern of spines, 9 = Longitudinal pattern of spines, SP = Special, T = Terminal. Taxa of subgenus Hyalochaete bear at least one rimoportula on terminal valves only. Evensen & Hasle's (1975) observations. It is hoped that the new findings and observations made in this study will be useful when changes or additions to the present classification are being considered. One new section is proposed: Peruviana, to include some heterovalvar species within subgenus Chaetoceros. The general classification, including the taxa studied and listed, plus those placed herein (*) in section Peruviana, is given below. Chaetoceros Ehrenb. Subgenus Chaetoceros (Phaeoceros Gran) I. Section Atlantica Ostenf. C. atlanticus var. neapolitanus (Schrod.) Hust. C. atlanticus var. skeleton (F. Schiitt) Hust. C. dichaeta Ehrenb. II. Section Borealia Ostenf. C. octagonus Hern.-Bec. C. densus Cleve C. tetrastichon Cleve C. seychellarus G. Karst. C. denticulatus Lauder C. rostratus Lauder C. danicus Cleve III. Section Peruviana Hern.-Bec. C. aequatorialis Cleve * C. convolutus Castrac. C. concavicornis L. Mangin C. criophilus Castrac. * C. curvatus Castrac. * C. peruvianus Brightw. C. pendulus G. Karst. IV. Section Coarctata Hern.-Bec. C. coarctatus Lauder Subgenus Hyalochaete Gran V. Section Dicladia (Ehrenb.) Gran C. decipiens Cleve C. decipiens f. singularis Gran C. lorenzianus Grunov C. lorenzianus f. forceps Meunier VI. Section Cylindrica Ostenf. C. bermejensis Hern.-Bec. VII. Section Compressa Ostenf. C. compressus Lauder VIII. Section Protuberantia Ostenf. emend. Hern.-Bec. C. didymus Ehrenb. C. protuberans Lauder IX. Section Constricta Gran C. constrictus Gran C. vanheurckii Gran X. Section Stenocincta Ostenf. 70 D. U. HERNANDEZ-BECERRIL C. affinis Lauder var. affinis C. affinis var. willei (Gran) Hust. C. paradoxus Cleve XI. Section Laciniosa Ostenf. C. distorts Cleve C. brevis F. Schutt XII. Section Diadema (Ehrenb.) Ostenf. emend. Gran C. diadema (Ehrenb.) Gran C. seiracanthus Gran XIII. Section Diversa Ostenf. C. diversus Cleve C. messanensis Castrac. XIV. Section Brevicatenata Gran C. pseudocrinitus Ostenf. C. wighami Brightw. XV. Section Curviseta Ostenf. emend. Gran C. curvisetus Cleve C. pseudocurvisetus L. Mangin C.debilis Cleve XVI. Section Anastomosantia Ostenf. C. rectus Hern. -Bee. XVII. Section Furcellata Ostenf. C. radicans F. Schutt C.filiferus G. Karst. XVIII. Section Socialia Ostenf. C. socialis Lauder var. ? XIX. Section Simplicia Ostenf. C. gracilis F. Schutt C. vistulae Apstein Subgenus Bacteriastroidea Hern. -Bee. C. bacteriastroides G. Karst. Biogeography Little effort has been made to assess the world distribution of Chaetoceros species, chiefly because many species have been recognized with uncertainty, while others are rather widespread or even cosmopolitan. Some species are, however, restricted to a given region or may be regarded as typically cold-water, warm-water or tropical forms, etc. The following account of general distribution patterns is based on information widely available in the literature, together with my own observations. The taxa are grouped according to the major regions. A. North or south cold-water ( occasionally found in temperate regions) C. adelianus Manguin C. atlanticus Cleve C. borealis Bailey C. bulbosus (Ehrenb.) Heiden C. castracanei G. Karst. C. concavicornis L. Mangin C. convolutus Castrac. C. criophilus Castrac. C. deflandrei Manguin C. dichaeta Ehrenb. C.flexuosus L. Mangin C. gausii Heiden & Kolbe C. hendeyi Manguin1 C. lawii Manguin C. natatus Manguin1 C. neglectus G. Karst. C. saltans Cleve ? B. Temperate to subtropical C. affinis Lauder var. affinis C. anastomosans Grunov C. brevis F. Schutt C. constrictus Gran C. curvisetus Cleve C. danicus Cleve C.debilis Cleve C. decipiens Cleve C. diadema (Ehrenb.) Gran C. didymus Ehrenb. C. distans Cleve C. eibenii Grunov ? C.filiferus G. Karst. C. gracilis F. Schutt C. laciniosus F. Schutt C. lorenzianus Grunov C. pelagicus Cleve C. protuberans Lauder C. pseudocurvisetus L. Mangin C. radicans F. Schutt C. socialis Lauder C. vistulae Apstein C. World-wide warm water (occasionally found in temperate regions) C. atlanticus var. neapolitanus (Schrod.) Hust. C. atlanticus var. skeleton (F. Schutt) Hust. C. compressus Lauder C. costatus Pavill. C. pendulus G. Karst.2 C. peruvianus Brightw.3 C. rostratus Lauder C. messanensis Castrac. D. Tropical and subtropical C. bacteriastroides G. Karst.? C. buceros G. Karst. C. coarctatus Lauder C. dadayi Pavill. C. denticulatus Lauder C. diversus Cleve C. paradoxus Cleve C. seychellarus G. Karst.2 C. tetrastichon Cleve 'Originally described from Antarctic waters (Manguin, 1960), with only one known record. 2Reported by Manguin ( 1 954) from a cold-water area. 3Reported by Sournia et al. (1979) from a temperate to cold-water area. ACKNOWLEDGEMENTS. Many thanks and appreciation are due to my supervisor, Dr R.M. Crawford for his encouragement and teaching during this investigation. I would like to thank Prof. F.E. Round, who offered very helpful and interesting discussions and advice. I also wish to thank Dr L.K. Medlin, Dr D.M. Paterson, Dr D.J. Patterson, Dr B. Leadbeater, Dr A. Beckett, and Prof. A. Walsby who provided comments and advice. Miss P. Sims and Dr D. Williams kindly provided encouragement and slides for reference. Acknowledgements are also due to Mr R. Porter and Mr S. Martin for their assistance with electron microscopy, and Mr T. Colborn for assistance in copying photographic plates. Dr P. Hargraves made comments on the work, and Mr B. Hartley provided some slides for reference. Most of the material was made available to the author by: M. en C.R. Cortes-Altamirano (ICMyL-UNAM), Dr S. Gomez-Aguirre (IB-UNAM), and the institutions CIB (Dr C. Lechuga-Deveze) and CICIMAR. Financial support was provided by CONACYT (Mexican Council for Science and Technology). STUDY OF CHAETOCEROS SPECIES REFERENCES Allen, W.E. & Cupp, E.E. 1935. Plankton diatoms of the Java Sea. Annls Jard. hot. Buitenz.44: 101-224. Andersen, R.A., Medlin, L.K. & Crawford, R.M. 1986. An investigation of the cell wall components of Actinocyclus subtilis (Bacillariophyceae). J. Phycol. 22: 466-479. Anonymous. 1975. Proposals for a standardization of diatom terminology and diagnoses. Beih. nov. Hedwigia 53: 323-354. Apstein, C. 1909. Chaetoceras gracile Schiitt und Chaetoceras vistulae n. sp. Wiss. Meeresunters.,Kielll: 133-137. Blasco, D. 1970. Estudio de la morfologia de Chaetoceros didymus al microscopic electronico. Investigation pesq. 34: 149-155. Brightwell, T. 1856. On the filamentous, long-horned diatomaceae, with a description of two new species. Q. Jlmicrosc. Sci. 4: 105-109, pi. VII. Brunei, J. 1966. Normalization de la terminologie des soies dans le genre Chaetoceros. Naturalist can. 93: 849-860. 1970. Le phytoplancton de la Bale des Chaleurs. 2nd ed. Montreal. 1972. Orientation of setae in the genus Chaetoceros, in regard to the apical axis. J. mar. biol. Ass. India 14( 1 ): 3 1 5-327. Castracane, F. 1875. Contribuzione alia florula delle diatomee del Mediterraneo ossia esame del contenuto nello stomaco di una salpa pinnata pescata a Messina. A tti A ccad. pontif. Nuovi Lincei 28: 377-396, pi. 6. 1886. Report on the diatomaceae collected by H.M.S. Challenger during the years 1873-1876. Rep. scient. Results Voy. Challenger (EoL) 2: 1-178, 30 pis. Cleve, P.T. 1 873a. Examination of diatoms found on the surface of the Sea of Java. Bih. K. svenska VetenskAkad. Handl. 1(1 1): 1-13, 3 pis. 1873ft. On diatoms from the Arctic Sea. Bih. K. svenska VetenskAkad. Handl. 1(1 3): 1-28, 4 pis. 1889. Pelagiske Diatomeer fran Kattegat. Vidensk. Udbytte Kanonbaaden 'Hauch's' Togterdanske Have. Part I. 2: 53-56. Copenhague. 1894. Redogorelse for de Svenska Hydrografiska undersokningar. Cilioflagellaten och Diatomaceer. Bih. K. svenska VetenskAkad. Handl. 20(3): 1-16, 2 pis. 1 897. A treatise of the phytoplankton of the A tlantic and its tributaries and on the periodical changes of the plankton ofSkagerak. Uppsala. 190 1 . The seasonal distribution of Atlantic plankton organisms. Goteborgs K. Vetensk. Vitter. Samh. Handl. IV, 3: 368 pp. Cleve-Euler, A. 1951. Die Diatomeen von Schweden und Finnland. K. svenska VetenskAkad. Handl. 2(1): 1-163. Crawford, R.M. 1975. The taxonomy and classification of the diatom genus Melosira C. Ag. I. The type species M. nummuloides C. Ag. Br. phycol. J. 10: 323-338. Gardner, C. & Medlin, L.K. 1994. The genus Attheya. I. A description of four new taxa, and the transfer of Gonioceros septentrionalis and G. armatus. Diatom Res. 9(1): 27-51. Cupp, E.E. 1943. Marine plankton diatoms of the west coast of North America. Bull. Scripps Instn Oceanogr. tech. Ser. 5(1): 1-238. Desikachary, T.V. 1956. Electron microscope studies on diatoms. Jl R. microsc. Soc. 111,76:9-36. & Bahadur, K. 1954. Electron microscope studies of diatom wall structure: II - Genus Chaetoceros Ehrenb. J. Sci. Industr. Res. 13 B: 92-94. Hema, A., Prasad, A.K.S.K., Sreelatha, P.M., Sridharan, V.T. & Suhrahmanyan, R. 1987. Marine diatoms from the Arabian Sea and Indian Ocean. Atlas of Diatoms, Fasc. IV: pis 332^00A. Madras. & Prema, P. 1987. Diatoms from the Bay of Bengal. Atlas of Diatoms, Fasc. Ill: pis 222-331. Drebes, G. 1974. Marines phytoplankton. Stuttgart. Duke, E.L., Lewin, J. & Reimann, B.E.F. 1973. Light and electron microscope studies of the diatom species belonging to the genus Chaetoceros Ehrenberg. I. Chaetoceros septentrionale0slrup. Phycologia 12: 1-9. Ehrenberg, C.G. 1 844. Einige vorlaufige Resultate der Untersuchungen der von der Sudpolreise des Captain Ross, so wie von den Herren Schayer und Darwin zugekommenen Materialen uber das verhalten der kleinsten Leben in den Oceanen und den grossten bisher zuganglichentiefen des Weltmeers. Ber. Akad. Wiss. Berlin 1844: 182-207. Evensen, D.L. & Hasle, G.R. 1975. The morphology of some Chaetoceros (Bacillariophyceae) species as seen in the electron microscopes. Beih. nov. Hedwigia 53: 153-174. Fryxell, G.A. 1978. Chain-forming diatoms: three species of Chaetoceraceae. J. Phycol. 14: 62-71. & Hasle, G.R. 1974. Coscinodiscineae: some consistent pattern in diatom morphology. Beih. nov. Hedwigia 45: 69-96. & Medlin, L.K. 1981 . Chain forming diatoms: evidence of parallel evolution in Chaetoceros. Cryptogamie Algol. 2(1): 3-29. Sims, P. A. & Watkins, T.P. 1986. Azpeitia (Bacillariophyceae): related genera and promorphology. Syst. Bot. Monogr. 13: 74 pp. 71 Giuffre, G. & Ragusa, S. 1988. The morphology of Chaetoceros rostratum L.auder (Bacillariophyceae) using light and electron microscopy. Botanica mar. 31: 503-510. Gran, H.H. 1 897. Protophyta: Diatomaceae, Silicoflagellata og Cilioflagellata. Den Norske Nordh.-Exp. (1876-78) 24: 1-36. 1904. Dia Diatomeen der arktischen meere. I Teil. Die Diatomeen des planktons. Fauna arct. 3(8): 59-554, 17 pis. 1905. Diatomeen. In K. Brandt & C. Apstein (Eds), Nord. Plankt. 19: 1-146. & Angst, E.C. 1 93 1 . Plankton diatoms of the Puget Sound. Publs Puget Sound mar. biol. Srn7:4l7-516. — & Yendo, K. 1914. Japanese diatoms. I. On Chaetoceras. II. On Stephanopyxis. Videnskap. Skr. I. Mat.-Naturv. Kl. 8: 1-20. Grunow, A. 1863. Ueber einige neue und ungenugend bekante Arten und Gattungen. Verh. zool.-bot. Ges. Wien 13: 137-162, pis 13, 14. Margraves, P.E. 1 972. Studies on marine plankton diatoms. I . Chaetoceros diadema (Ehr.) Gran: life cycle, structural morphology, and regional distribution. Phycologia 11: 247-257. 1979. Studies on marine plankton diatoms. IV Morphology of Chaetoceros resting spores. Beih. nov. Hedwigia 64: 99-120. Hasle, G.R. 1978. Diatoms. In A. Sournia (Ed.), Phytoplankton manual: 136-142. Paris. Helmcke, J.-G. & Krieger, W. 1953. Diatomeenschalen im elektronenmikroscopischen Bild. 1. Winheim. 1954. Diatomeenschalen im elektronenmikroscopischen Bild. 2. Winheim. Hendey, N.I. 1 937. The plankton diatoms of the Southern Seas. 'Discovery' Rep. 16: 151-364. 1959. The structure of the diatom cell wall as revealed by electron microscope. J. Quekett microsc. Club IV, 5: 147-175. 1964. An introductory account of the smaller algae of British coastal waters. Part V. Bacillariophyceae. Fish. Inv. ser. IV London. Ctishing, D.H. & Ripley, G.W. 1954. Electron microscope studies of diatoms. Jl R. microsc. Soc. Ill, 74: 22-34. Hernandez-Becerril, D.U. 1 99 1 a. The morphology and taxonomy of the planktonic diatom Chaetoceros coarctatus Lauder (Bacillariophyceae). Diatom Res. 6(2): 281-287. 19916. Note on the morphology of Chaetoceros didymus and C. protuberans, with some considerations on their taxonomy. Diatom Res. 6(2): 289-297. 199 lc. Chaetoceros bermejensis sp. nov., a new planktonic diatom from the Gulf of California. Botanica mar. 34: 521-526. 1992a. Observations on two closely related species, Chaetoceros tetrastichon and C. t&doy/ (Bacillariophyceae). Nord. J. Bot. 12: 365-371. 1992ft. Two new species of the diatom genus Chaetoceros (Bacillariophyta). Pl.Syst.Evol. 181:217-226. 1993a. Note on the morphology of two planktonic diatoms: Chaetoceros bacteriastroides and C. seychellarus, with comments on their taxonomy and distribution. Bot. J. Linn. Soc. Ill: 1 17-128. 1993ft. Study of the morphology and distribution of two planktonic diatoms: Chaetoceros paradoxus and Ch. filiferus (Bacillariophyceae). Crypt. Bot. 3: 169-175. Meave del C., M.E. & Lara Villa, M.A. 1993. Observations on Chaetoceros buceros (Bacillariophyceae), a rare tropical planktonic species collected from the Mexican Pacific. J. Phycol. 29: 811-818. Hustedt, F. 1920-1921. Atlas der Diatomaceen-Kunde, pis 321-327, 337-344. In A. Schmidt, At las der Diatomaceen-Kunde. Leipzig. 1930. Die Kieselalgen Deutschlands, Osterreichs und der Schweiz. In G.L. Rabenhorst, Kryptogamen-Floral(\): 1-920. Ikari, J. 1928. On some Chaetoceras of Japan. II. Bot. Mag., Tokyo 42(497): 247-262. Karsten, G. 1905. Das Phytoplankton der Antarktischen Meeres nach dem material der deutschen Tiefsee-Expedition 1898-1899. Wiss. Ergebn. dt. Tiefsee-Exped. ' Valdivia' II, 2(1): 1-136, pis 1-19. 1907. Das Indischen Phytoplankton. Wiss. Ergebn. dt. Tiefsee-Exped. ' Valdivia II, 2(3): 221-548, pis 35-54. Koch, P. & Rivera, P. 1 984. Contribution al conocimiento de las diatomeas chilenas. III. El genero Chaetoceros Ehr. (subgenero Phaeoceros Gran). Gayana Bot. 41(1-2): 61-84. Krasske, G. 1941. Die Kieselalgen des chilenischen kustenplanktons (aus dem sudchilenischen kustengebiet, Beitrag a). Arch. Hydrobiol. 38(2): 260-287, pis 4-6. Lauder, H.S. 1 864. Remarks on the marine diatomaceae found at Hong Kong, with description of new species. Trans, microsc. Soc., London, N.S., 12: 75-79, pi. VIII. Lebour,M.V. 1930. The plank tonic diatoms of northern seas. London. Lechuga-Deveze, C.H. & Hernandez-Becerril, D.U. 1988. Life cycle of the diatom Chaetoceros protuberans Lauder (1864) (Bacillariophyceae). Investigacion pesq. 52(1): 77-83. Leger, G. 1973. Diatomees et Dinoflagelles de la cote Est de Corse. Systematique et distribution en juillet 1964. Bull. Inst. oceanogr. Monaco 71 (1426): 1-31. 72 D. U. HERNANDEZ-BECERRIL Leuduger-Fortmorel, G. 1892. Diatomees de la Malaisie. Annls Jard. hot. Buitenz. 11: 1-60, 7 pis. Li, C.-W. & Volcani. B.E. 1985. Studies on the biochemistry and fine structure of silica shell formation in diatoms. IX. Sequential valve formation in a centric diatom, Chaetocerosrostratum. Protoplasma 124: 30-41. Mangin, L. 1910. Sur quelques algues nouvelles ou peu connues du phytoplankton del'Atlantique. Bull. Soc. hot. Fr. 57: 344-350. — 1917. Sur le Chaetoceros criophilus Castr., espece caracteristique des mers antarctiques. C. r. hebd. Seanc. Acad. Sci., Paris 164: 704-709. — 1919. Sur les Chaetoceros du groupe Peruvianus Bgtw. Bull. Mus. natn. Hist, not., Paris 25(4, 5): 305-310, 41 1^14. Manguin, E. 1954. Diatomees marines de 1'lle Heard (Australian Antarctic Research Expedition). Rev. algol. N.S., 1(1): 14-24. — 1960. Les diatomees de la Terre Adelie, campagne du 'Commandant Charcot' 1949-1950. Annls Sci. not. (Bot.) XII, 1(2): 223-363, incl. 31 pis. Marino, D., Giuffre, G., Montresor, M. & Zingone, A. 1991. An electron microscope investigation on Chaetoceros minimus (Levander) comb. nov. and new observations on Chaetoceros throndsenii (Marino, Montresor and Zingone) comb. nov. Diatom Res. 6(2): 317-326. — Montresor, M. & Zingone, A. 1987. Miraltia throndsenii gen. nov., sp. nov., a planktonic diatom from the Gulf of Naples. Diatom Res. 2(2): 405-41 1. Medlin, L.K., Crawford, R.M. & Andersen, R.A. 1986. Histochemical and ultrastructural evidence for the function of the labiate process in the movement of centric diatoms. Br.phycol. J. 21: 297-301. Meunier, A. 1913. Microplancton de la Mer Flamande. 1. Le genre Chaetoceros Ehr. Mem. Mus. R. Hist. not. Belg. 7(2): 1-58, pis 1-7. Moreno Ruiz. J.L., Soto, P.J., Zamudio, M.E., Hernandez-Becerril, D.U. & Licea D., S. 1993. Morphology and taxonomy of Chaetoceros diversus (Bacillariophyceae) based on material from the Southern Gulf of Mexico. Diatom Res. 8: 419-428. Navarro, J.N. 1982. A survey of the marine diatoms of Puerto Rico. HI. Suborder Biddulphiineae: Family Chaetoceraceae. Botanicamar. 25: 305-319. Okamura, K. 1907. Some Chaetoceras and Peragallia of Japan. Bot. Mag., Tokyo 31(144): 89-106, pis 3, 4. Okuno, H. 1951. Electron microscopical study on antarctic diatoms (1). J. Jap. Bot. 26(10): 305-310. — 1956. Electron microscopical study on fine structures of diatom frustules. XIV. Observations on the genus Chaetoceros. Bot. Mag., Tokyo 69(14): 1 82-200. — 1970. Marine diatoms. In J.G. Helmcke& W. Krieger(Eds), Diatomeenschalen im elektronmikroskopischen Bild. 7: 31 pp., pis 614-714. Lehre. Ostenfeld, C.H. 1901. lagttagelser over Plankton-Diatomeer. Nyt Mag. Naturvid. 39(4): 287-302. — 1903. Phytoplankton from the sea around the Faeroes. In E. Warming, Botany of the Faeroes 2: 558-612. Copenhagen. 1912. A revision of the marine species of Chaetoceros Ehbg. sect. Simplicia Ostf. Meddr Kommn Havunders. (Plankton) 1(10): 1-11. Peragallo, H. & Peragallo, M. 1897-1908. Diatomees marines de France et des districts maritimes voisins. Grez-sur-Loing. Ratkova, T.N. 1974. The morphological features and geographical distribution of Chaetoceros chilensis Krasske and Ch. peruvianus Brightwell (Bacillariophyta). Bot. Zh. SSSR 59: 66-74, 2 pis [In Russian]. Rines, J.E.B. & Margraves, P.E. 1986. Considerations of the taxonomy and biogeography of Chaetoceros ceratoporosus Ostf. and Chaetoceros rigidus Ostf. In M. Ricard (Ed.), Proc. 8th Symp. Dial.: 97-1 12. - 1988. The Chaetoceros Ehrenberg (Bacillariophyceae) flora of Narragansett Bay, Rhode Island, U.S.A. Biblthca phycol. 79: 196 pp. 1990. Morphology and taxonomy of Chaetoceros compressus var. hirtisetus var. nova, with preliminary consideration of closely related taxa. Diatom Res. 5: 113-127. Rogerson, A., De Freitas, A.S.W. & Mclnnes, A.G. 1986. Growth rates and ultrastructure of siliceous setae of Chaetoceros gracilis (Bacillariophyceae). J. Phycol. 22(1): 56-62. Ross, R., Cox, E.J., Karayeva, N.I., Mann, D.G., Paddock, T.B.B., Simonsen, R. & Sims, P. A. 1979. An amended terminology for the siliceous components of the diatom cell. Beih. nov. Hedwigia64: 513-533. Round, F.E. & Crawford, R.M. 1981. The lines of evolution of the Bacillariophyta. I. Origin. Proc. R. Soc. B, 211: 237-260. & Mann,D.G. 1990. The Diatoms. Biology and morphology of the genera. Cambridge. Schroder, B. 1900. Das phytoplankton des Golfes von Neapel nebst vergleichenden Ausblicken auf das atlantischen Ocean. Mitt. zool. Stn Neapel 14: 1-38, pi. 1. Schiitt, F. 1895. Arten von Chaetoceras und Peragallia. Ein Beitrag zur Hochseeflora. Ber. dt. hot. Ges. 13: 35^8, pis IV, V. Simonsen, R. 1974. The diatom plankton of the Indian Ocean expedition of R.V. 'Meteor' 1964-1965. Meteor ForschErgebn. D, 19: 1-66,41 pis. 1979. The diatom system: ideas on phylogeny. Bacillaria 2: 9-71 . Sournia, A. 1968. Diatomees planctoniques du Canal de Mozambique et de File Maurice. Mem. O.R.S. TO. M. 31: 1-120, 13 pis. Grail, J.-R. & Jacques, G. 1979. Diatomees et dinoflagelles planctoniques d'une coupe meridienne dans le sud de 1'Ocean Indien (campagne 'Antiprod I' du Marion-Dufresne, mars 1977). Botanicamar. 22: 183-198. Stockwell, D.A. & Hargraves, P.E. 1986. Morphological variability within resting spores of the marine diatom genus Chaetoceros Ehrenberg. In M. Ricard (Ed.), Proc. 8th Symp. Dial.: 81-95. Stosch, H.A. von. 1977. Observations on Bellerochea and Streptotheca, including descriptions of three new planktonic diatom species. Beih. nov. Hedwigia 54: 113-166. — Theil, G. & Kowallik, K.V. 1973. Entwicklungsgeschichtliche Untersuchungen an zentrichen Diatomeen. V. Bau und Lebenszyklus con Chaetoceros didymum, mit Beobachtungen uber einige andere Arten der Gattung. Helgolander wiss. Meeresunters. 25: 384—445. Suhrahmanyan, R. 1946. A systematic account of the marine plankton diatoms of the Madras coast. Proc. Indian Acad. Sci. B, 24(4): 85-197. Takano, H. 1983. New and rare diatoms from Japanese marine waters. X. A new Chaetoceros common in estuaries. Bull. Tokai reg. Fish. Res. Lab. 110: 1-11. Taylor, F.J.R. 1966. Phytoplankton of the south western Indian Ocean. Nova Hedwigia 12(3-4): 433^176, pis 88-96. VanLandingham, S.L. 1 968. Catalogue of the fossil and recent genera and species of diatoms and their synonyms. 2. Lehre. STUDY OF CHAETOCEROS SPECIES 73 INDEX Acanthoceros Honigm. 2, 68 Attheya West 2, 68 Bacteriastrum Shadbolt 2, 68 Chaetoceros adelianus Manguin 70 aequatorialis Cleve 16 affinis Lauder var. affinis 35, 36, 37 affinis var. circinalis (Meunier) Hust. 38 affinis var. willei (Gran) Hust. 38, 39 amanita A. Cleve 53 anastomosans Grunov 70 atlanticus Cleve 70 atlanticus var. neapolitanus (Schrod.) Hust. 4, 6, 7 atlanticus var. skeleton (F. Schiitt) Hust. 4, 8, 9 bacteriastroides G. Karst. 67 bermejensis Hern. -Bee. 32, 67 borealis Bailey 10 borealis var. densa Cleve 1 0 brevis F. Schutt 38, 42, 43 buceros G. Karst. 67 bulbosus (Ehrenb.) Heiden 70 castracanei G. Karst. 70 chilensis Krasske 22 coarctatus Lauder 27, 67 compressus Lauder 32, 33, 34 compressus var. hirtisetus Rines & Hargraves 67 concavicornis L. Mangin 21 concaviformis L. Mangin 21 constrictus Gran 35, 39 convolutus Castrac. 21 costatus Pavill. 70 crinitus F. Schutt 53 criophilus Castrac. 2 criophilus Gran 2 1 curvatus Castrac. 16 curvisetus Cleve 53, 54, 55 dadayi Pavill. 67 danicus Cleve 16, 19,20 debilis Cleve 58, 59, 60 Accepted names are in roman and synonyms in italic. decipiens Cleve 27, 28, 29 decipiens f. singularis Gran 27, 32 deflandrei Manguin 70 densusClevelO,13, 14 denticulatus Lauder 10, 15 denticulatus f. angusta Hust. 10 denticulatus f. lata Hust. 10 diadema (Ehrenb.) Gran 38, 44, 45 dichaeta Ehrenb. 8, 11,12 didymus Ehrenb. 35 distansCleve38,40,41 diversus Cleve 46, 47, 48, 49, 50 eibenii Grunov 70 filiferus G. Karst. 58 flexuosus L. Mangin 70 furca Cleve 46 gausii Heiden & Kolbe 70 glandaziiL. Mangin 16 gracilis F. Schutt 55, 63 hendeyi Manguin 70 ingolfianus Ostenf. 53 janischianus Castrac. 8 laciniosus F. Schutt 38, 70 laevis Leud.-Fortm. 46 lawii Manguin 70 lorenzianus Grunov 27, 30, 3 1 lorenzianus f. forceps Meunier 31, 32 messanensis Castrac. 46, 5 1 , 52 nanodenticulatus O\i&mwd 10 natatus Manguin 70 neapolitanus Schrod. 4 neglectus G. Karst. 70 octagonus Hern. -Bee. 10 paradoxus Cleve 38 pelagicus Cleve 70 pendulus G. Karst. 22, 25, 26 peruvianus Brightw. 22, 23, 24 peruvianus f. gracilis (Schrod.) Hust. 22 peruvianus f. robusta (Cleve) Hust. 22 polygonus F. Schutt 4 protuberans Lauder 35 pseudocrinitus Ostenf. 53 pseudocurvisetus L. Mangin 53, 56, 57 radicans F. Schutt 58, 61, 62 rectus Hern.-Bec. 58 rostratus Lauder 16, 17, 18 rostratus var. glandazii(L. Mangin) F.J.R. Taylor 16 rudis Leud.-Fortm. 68 salsugineus Takano 2 saltans Cleve 70 schuetti Cleve 38 secundus Cleve 53 seiracanthus Gran 43 seychellarus G. Karst. 10 skeleton F. Schutt 4 socialis Lauder 63 socialis Lauder var. ? 63, 64, 65 subsecundus (Grunov) Hust. 43 tetrastichon Cleve 10, 67 vanheurcki Gran 35, 39 vistulae Apstein 63, 66 wighami Brightw. 53, 55 willei Gran 38 Dicladia Ehrenb. 2 Gonioceros H. Perag. & Perag. 2, 68 armatus (West) Peragallo 68 septentrionalis (0strup) R.M. Crawford 68 Goniothecium Ehrenb. 2 Leptocylindrus Cleve 68 Miraltia D. Marino, Montresor & Zingone 68 Peragallia F. Schutt 2 Syndendrium Ehrenb. 2 Bulletin of The Natural History Museum Botany Series Earlier Botany Bulletins are still in print. The following can be ordered from Intercept (address on inside front cover). Where the complete backlist is not shown, this may also be obtained from the same address. Volume 1 No. 4 A revision of the Macaronesian genus Argyranthemum Webb No. 1 Contributions to our knowledge of Old World Araliaceae. ex Schultz Bip. (Compositae-Anthemideae).C.J. Humphries. 1 976. Pp. 1 45-240, 2 plates, 26 figs. £7.25 No. 5 Frank Ludlow (1885-1972) and the Ludlow-Sherriff expeditions to Bhutan and south-eastern Tibet of 1933-1950. W.T. Stearn. 1976. Pp. 243-268, 1 fig. Reliquiae botanicae himalaicae. F. Ludlow. 1976. Pp. 269-289, 7 plates, 8 figs. £5.25 Contributions to our knowledge of Old World Araliaceae. W.R. Philipson. 1951. Pp. 3-20. An undescribed species of Mastichodendron (Sapotaceae) from Barbados and Antigua. H.E. Box & W.R. Philipson. 1951. Pp. 2 1-24,1 plate. £2.80 No. 2 Out of print. No. 3 Notes on Podostemaceae for the revision of the Flora of West Tropical Africa. G. Taylor. 1 953 . Pp. 5 1 -79, 1 4 figs. £2.15 No. 4 Notulae criticae and floram hispaniae pertinentes, 1 . V.H. Heywood. 1954. Pp. 81-122. £2.95 Nos 5-8 Out of print. Volume 2 No. 1 Out of print. No. 2 Out of print. No. 3 Novitates Himalaicae-1 . F. Ludlow & W.T. Stearn. 1 956. Pp. 65-8 1,8 plates, 11 figs. £2.15 No. 4 Saxifraga of the Himalaya 1 . Section Kabschia. H. Smith. 1958. Pp. 83-129, 14 figs. £4.50 Nos 5- 1 2 Out of print. Volume 3 Nos 1-3 No. 4 No. 5 No. 6 Volume 4 Nos 1-3 No. 4 No. 5 No. 6 No. 7 No. 8 £1.95 Out of print. A revision of the genus Petrorhagia. P.W. Ball & V.H. Heywood. 1964. Pp. 1 19-172, 3 plates, 22 figs. Marine algae of Gough Island. Y.M. Chamberlain. 1965. Pp. 1 73-232, 4 plates, 80 figs. £6.00 Out of print. Out of print. A synopsis of Jamaican Myrsinaceae. W.T. Stearn. 1969. Pp. 143-1 78, 8 plates, 25 figs. £3.65 The Jamaican species of Columnea and Alloplectus (Gesneriaceae). W.T.Stearn. 1969. Pp. 179-236, 8 plates, 29 figs. £5.15 New or little known Himalayan species of Swertia and Veratrilla (Gentianaceae). H. Smith. 1970. Pp. 237-258, 16 plates, 7 figs. £3.75 A survey of the tropical genera Oplonia and Psilanthele (Acanthaceae). W.T. Stearn. 1971. Pp. 259-323, 10 plates, 18 figs. £5.95 Angiosperms of the islands of the Gulf of Guinea (Fernando Po, Principe, S. Tome, and Annobon). A.W. Exell. 1973. Pp. 325-411. £6.00 Volume 5 No. 1 The dryopteroid ferns of Ceylon. W. A. Sledge. 1 973. Pp. 1-43, 4 figs. £3.00 No. 2 New Himalayan and Tibetan species of Corydalis (Papaveraceae). F. Ludlow & W.T. Stearn. 1975. Pp. 45-69, 15 plates, 14 figs. £3.85 No. 3 The marine algae of Trinidad, West Indies. W.D. Richardson. 1975. Pp. 71-143, 12 plates, 2 figs. £6.80 No. 6 No. 7 Volume 6 No. 1 No. 2 No. 3 No. 4 Studies in the genus Hypericum L. (Guttiferae). 1 . Infrageneric classification. N.K.B. Robson. 1977. Pp. 291-355, 9 figs. £7.25 Sphagnales of tropical Asia. A.Eddy. 1977. Pp. 357^45, 4 plates, 1 7 maps, 25 figs. £9.25 The handwriting of Joseph Banks, his scientific staff and amanuenses. J.B. Marshall. 1978. Pp. 1-85, 62 figs. £9.50 Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. II. Phaeophyta. J.H. Price, D.M. John & G.W Lawson. 1978. Pp. 87-182, 1 fig. £15.00 The lichenicolous Hyphomycetes. D.L. Hawksworth. 1979. Pp. 183-300, 47 figs. £15.00 The species of Chisocheton (Meliaceae). D.J. Mabberley. 1979. Pp. 301-386, 3 plates, 10 figs. £15.00 Volume 7 No. 1 No. 2 No. 3 Volumes No. 1 No. 2 No. 3 No. 4 The distribution of Padina pavonica (L.) Lamour. (Phaeophyta: Dictyotales) on British and adjacent European shores. J.H. Price, I. Tittley & W.D. Richardson. 1979. Pp. 1-67, 3 plates, 2 figs. £9.00 Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. 111. Rhodophyta (Bangiophyceae). D.M. John, J.H. Price, C.A. Maggs, G.W. Lawson. 1979. Pp. 69-82, 1 fig. £2.20 A revision of the genus Anacyclus L. (Compositae: Anthemideae). C.J. Humphries. 1979. Pp. 83-142, 27 figs. £8.00 The Thelypteridaceae of Ceylon. WA. Sledge. Pp. 1-54, 5 figs. 1981. £7.75 Studies in the genus Hypericum L. (Guttiferae) 2. Characters of the genus. N.K.B. Robson. 1981 . Pp. 55-226, 73 figs. £23.50 A revision of the lichen family Thelotremataceae in Sri Lanka. M.E. Hale, Jr. 1981 . Pp. 227-332, 20 figs. £14.50 Vascular plant collections from the Tristan da Cunha group of islands. E.W. Groves. Pp. 333^20, 33 figs. £12.50 Volume 9 No. 1 The lichenicolous Coelomycetes. D.L. Hawksworth. 1981 . Pp. 1-98, 36 figs. £13.25 No. 2 The genus Callithamnion (Rhodophyta: Ceramiaceae) in the British Isles. PS. Dixon & J.H. Price. 1981. Pp. 99-141, 5 figs. £6.00 No. 3 Parmelia subgenus Amphigymnia (lichens) in East Africa. H . Krog & T.F.V. Swinscow. 1981 . Pp. 143-231 , 31 figs. £12.25 No. 4 The genus Selaginella in tropical South America. A.H.G. Alston, A.C. Jermy & J.M. Rankin. 1981 . Pp. 233-330, 18 figs. £13.50 Volume 10 No. 1 Taxonomic studies in the Labiatae tribe Pogostemoneae. J.R. Press. 1982. Pp. 1-89, 33 figs. £12.75 No. 2 The typification of Hudson's algae: a taxonomic and nomenclatural reappraisal. L.M. Irvine & PS. Dixon. 1982. Pp. 91-105. £2.25 No. 3 Seaweeds of the Faroes. Various authors. 1982. Pp. 107-225, 13 figs. £16.75 No. 4 The lichen genus Steinera. A.M. Henssen & P.W. James. 1 982. Pp. 227-256, 24 figs. £4.50 Volume 11 No. 1 The algae of Lightfoot's Flora scotica. P.S. Dixon. 1983. Pp. 1-1 5, 2 figs. £2.30 No. 2 A taxonomic study of the lichen genus Micarea in Europe. B.J. Coppins. 1983. Pp. 17-214, 57 figs, 28 maps. £27.75 No. 3 The hepatics of Sierra Leone and Ghana. E.W. Jones & A.J. Harrington. 1983. Pp. 215-289, 8 figs. £10.50 No. 4 Studies in the Corallinaceae with special reference to Fosliella and Pneophyllum in the British Isles. Y.M. Chamberlain. 1983. Pp. 291^463, 89 figs. £24.50 Volume 12 No. 1 A revision of the Morinaceae (Magnoliophyta-Dipsacales). M.J. Cannon & J.F.M. Cannon. 1984. Pp. 1-35, 9 figs. £5.40 No. 2 An introduction to fern genera of the Indian subcontinent. C.R. Eraser- Jenkins. 1984. Pp. 37-76, 1 fig. £6.00 No. 3 A revision of African Sphagnales. A. Eddy. 1985. Pp. 77- 162, 47 figs. £13.50 No. 4 Studies in the genus Hypericum L. (Guttiferae) 3. Sections 1 . Campylosporus to 6a. Vmbraculoides. N.K.B. Robson. 1985. Pp. 1 63-325, 24 figs, 34 maps. £28.80 Volume 13 No. 1 The lichen genus Usnea subgenus Neuropogon. F.J. Walker. 1985. Pp. 1-130, 39 figs. £19.80 No. 2 Cytotaxonomic studies of the ferns of Trinidad. A.C. Jermy &T.G. Walker. 1985. Pp. 131-276, 69figs. £22.00 No. 3 Some genera of the Biddulphiaceae (diatoms) with interlocking linking spines. R. Ross & PA. Sims. 1985. Pp. 277-38 1,33 plates. £20.00 Volume 14 No. 1 Cytological observations on Indian subcontinent and Chinese Dryopteris and Polystichum (Pteridophyta: Dryopteridaceae). M. Gibby. 1985. Pp. 1^42, 78 figs. £6.00 No. 2 A redisposition of the species referred to the ascomycete genus Microthelia. D.L. Hawksworth. 1985. Pp. 43-181, 73 figs. £24.00 No. 3 A classification of the genus Dryopteris (Pteridophyta: Dryopteridaceae). C.R. Fraser- Jenkins. 1986. Pp. 183-218,4 figs. £5.60 No. 4 Evolutionary cladistics of marattialean ferns. C.R. Hill & J. M . Camus. 1 986. Pp. 2 1 9-300, 27 figs. £1 4.50 Volume 15 No. 1 Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 1 . Genera A-F. J.H. Price, D.M. John & G.W. Lawson. 1986. Pp. 1-122, 1 fig. £23.50 No. 2 Cytology of the fern flora of Madeira. I. Manton, J.D. Lovis, G. Vida& M. Gibby. 1986. Pp. 123-161, 12plates. £7.50 No. 3 A revision of the lichen genus Xanthoparmelia in Australasia. J.A. Elix, J. Johnston & P.M. Armstrong. 1986. Pp. 1 63-362, 42 figs, 1 1 7 maps. £38.00 Volume 16 No. 1 Studies in the genus Hypericum L. (Guttiferae) 7. Section 29. Brathys(part 1). N.K.B. Robson. 1987. Pp. 1-106, 14 figs, 25 maps. £20.00 No. 2 The lichen genus Ramalina in Australia. G.N. Stevens. 1987. Pp. 1 07-233, 1 5 plates, 3 1 figs. £22.50 No. 3 An annotated list of vascular plants collected in the valleys south of Mt Everest. G. Miehe. 1987. Pp. 225-268, 4 figs. £8.50 No. 4 Further genera of the Biddulphiaceae (diatoms) with interlocking linking spines. R. Ross & A.Sims. 1987. Pp. 269-3 11, 13 plates. £8.00 Volume 17 No. 1 Studies in Pseudocyphellaria (lichens) 1 . The New Zealand species. D.J. Galloway. 1988. Pp. 1-267, 124 figs. £51.00 Volume 18 No. 1 An illustrated catalogue of the type specimens in the Greville diatom herbarium. D.M. Williams. 1988. Pp. 1-148, 74 plates. £28.00 No. 2 Erik Acharius and his influence on English lichenology. D.J. Galloway. 1 988. Pp. 149-1 94, 1 8 figs. £8.80 No. 3 Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 2. Genera G. J.H. Price, D.M. John & G.W. Lawson. 1988. Pp. 195-273, 1 fig. £14.80 No. 4 Some Cretaceous and Palaeogene Trinacria (diatom) species. PA. Sims & R. Ross. 1988. Pp. 275-322, 13 plates. £9.10 No. 5 A monograph of Dryopteris (Pteridophyta: Dryopteridaceae) in the Indian subcontinent. C.R. Fraser- Jenkins. 1 989. Pp. 323-477, 79 figs. £30.00 No. 6 Corydalis (Papaveraceae: Fumarioideae) in Nepal. M. Liden. 1 989. Pp. 479-538, 26 figs. £1 1 .20 Volume 19 A new species of Maytenus (Celastraceae) in Ethiopia. Sebsebe Demissew. 1989. Pp. 1-3, 1 fig. Central American Araliaceae - a precursory study for the Flora Mesoamericana. M.J. Cannon & J.F.M.Cannon. 1989. Pp. 5-6 1,36 figs. A revision of the Solatium nitidum group (section Holophylla pro parte): Solanaceae. S. Knapp. 1989. Pp. 63-102, 21 figs. Six new species of Solatium sect. Geminata from South America. S. Knapp. 1989. Pp. 103-1 12, 8 figs. The application of names of some Indian species of Ocimum and Geniosporum (Labiatae). J.R. Press & V.V. Sivarajan. 1989. Pp. 11 3-1 16, 4 figs. Revision of Piper (Piperaceae) in the New World 1 . Review of characters and taxonomy of Piper section Macrostachys. M.C. Tebbs. 1989. Pp. 1 17-158, 41 figs. £48.00 Volume 20 No. 1 Studies in the genus Hypericum L. (Guttiferae) 8. Sections 29. Brathys (part 2) and 30. Trigynobrathys. N.K.B. Robson. 1990. Pp. 1-151, 22 figs, 46 maps. £45.00 No. 2 The marine algal flora of Namibia: its distributions and affinities. G.W. Lawson, R.H. Simons and WE. Isaac. 1990. Pp. 153-168, 1 fig, 7 plates. The infrageneric classification of Gentiana (Gentianaceae). T.-N. Ho and S.-W Liu. 1990. Pp. 169-192, 13 figs. Revision of Piper (Piperaceae) in the New World. 2. The taxonomy of Piper section Churumayu. M.C. Tebbs, 1990. Pp. 1 93-236, 49 figs. £25.00 Volume 21 No. 1 Historical and taxonomic studies in the genus Titanoderma (Rhodophyta, Corallinales) in the British Isles. Y.M. Chamberlain. 1991 . Pp. 1-80, 247 figs. No. 2 Early collections of the Holy Thorn ( Crataegus monogyna cv. Biflora). A.R. Vickery. 1991. Pp. 81-83, 1 fig. A taxonomic study of the species referred to the ascomycete genus Leptorhaphis. B. Aguirre-Hudson. 1991. Pp. 85-192, 76 figs. The typification and identification of Calymperes crass ilimbatum Renauld & Cardot (Musci: Calymperaceae). L.T. Ellis. 1991. Pp. 193-194, 1 fig. Volume 22 No. 1 An account of southern Australian species of Lithophyllum (Corallinaceae, Rhodophyta). Wm. J. Woelkerlingand S.J. Campbell. 1992. Pp. 1-107, 63 figs. £37.00 No. 2 Palynological evidence for the generic delimitation of Sechium (Cucurbitaceae) and its allies. J.L. Alvarado, R. Lira-Saade&J.Caballero. 1992. Pp. 109-121. Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 3. Genera H-K. J.H. Price, D.M. John & G.W. Lawson. 1992. pp. 123-146. Two new species of Solatium section Geminata (Solanaceae) from Cerro del Torra in western Colombia. S. Knapp. 1992. Pp. 147-152. Fissidens ceylonensis Dozy & Molkenb. (Musci: Fissidentaceae) and some allied taxa from southern India. L.T. Ellis. 1992. Pp. 153-156, 2 figs. New species of Piper (Piperaceae) from Central America. M.Tebbs. 1992. Pp. 157-158. Studies on the Cretan flora 1 . Floristic notes. N.J. Turland. 1992. Pp. 159-164. Studies on the Cretan flora 2. The Dianthus juniperinus complex (Caryophyllaceae). N.J. Turland. 1992. Pp. 165-169. £37.50 Volume 23 No. 1 Revision of Piper (Piperaceae) in the New World 3. The taxonomy of Piper sections Lepianthes and Radula. M.C. Tebbs. 1993. Pp. 1-50, 18 figs. Mounting techniques for the preservation and analysis of diatoms. S.J. Russell. 1 993 . Pp. 5 1 -54. 1 fig. £37.50 No. 2 New taxa of Gentiana (Gentianaceae) from Western China and the Himalayan region. T.-N. Ho and S.-W. Liu. 1993. Pp. 55-60. 2 figs. New combinations, names and taxonomic notes on Genlianella (Gentianaceae) from South America and New Zealand. T.-N. Ho and S.-W Liu. 1993. Pp. 61-66. Studies in Hypericum: validation of new names. N.K.B. Robson. 1993. Pp. 67-70. Generic monograph of the Asteraceae-Anthemideae. K. Bremer and C.J. Humphries. 1993. Pp. 71-177. 12 figs. £37.50 Volume 24 No. 1 Pre-Linnaean references for the Macaronesian flora found in Leonard Plukenet's works and collections. J. Francisco-Ortega, A. Santos-Guerra and C.E. Jarvis. Pp. 1-34. Studies on the lichen genus Sticta (Schreber) Ach.: II. Typification of taxa from Swartz's Prodromus of 1788. D.J. Galloway. Pp. 35-48. Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 4. Genera L O. D.M. John, G.W. Lawson, J.H. Price, WF. Prud'homme van Reine and W.J. Woelkerling. Pp. 49-90. Studies on the Cretan flora 3. Additions to the flora of Karpathos. N.J. Turland and L. Chilton. Pp. 91-100. No. 2 Observations on the benthic marine algal flora of South Georgia: a floristic and ecological analysis. D.M. John, P.J.A. Pugh and I. Tittley. Pp. 101-1 14. Studies in Pseudocyphellaria (Lichens) IV. Palaeotropical species (excluding Australia). D.J. Galloway. Pp. 1 15-160. Morphology and ecology of seedlings, fruits and seeds of Panama: Bixaceae and Cochlospermaceae. N.C. Garwood. Pp. 161-172. A study of Bixa (Bixaceae), with particular reference to the leaf undersurface indumentum as a diagnostic character. R.E. Dempsey and N.C. Garwood. Pp. 173-180. Volume 25 No. 1 A revision of Rutilaria Greville (Bacillariophyta). R. Ross. Pp. 1-94. William Roxburgh's St Helena plants. Q.C.B. Cronk. Pp. 95-98. No. 2 Seaweeds of the western coast of tropical Africa and adjacent islands: a critical assessment. IV. Rhodophyta (Florideae) 5. Genera P. G.W. Lawson, W.J. Woelkerling, J.H. Price, W.F. Prud'homme Van Reine and D.M. John. Pp. 99-122. A new species of Odontorrhynchos (Orchidaceae, Spiranthinae) from Boliva.D.L. Szlachetko. Pp. 123-125. Linnaeus's interpretation of Prospero Alpine's De plant is exoticis, with special emphasis on the flora of Crete. N.J. Turland. Pp. 127-159. Book review. M.G. Gilbert. P. 161. CONTENTS A morphological study of Chactoccros species (Bacillariophyta) from the plankton of the Pacific Ocean of Mexico D.U. Hernandez-Becerril jlletin of The Natural History Musei BOTANY SERIES Vol.26, No. 1, June 1996