Chaetoceros mitra

This was the first species of the C. lorenzianus complex to be described, as Dicladia mitra from the Sea of Kamtschatka (Fig 20A) [17] based only on a valve of a resting spore. When chains of cells were found, the taxon was transferred to Chaetoceros [18]. Cleve described the cells as forming straight chains with narrow peanut-shaped to narrow elliptical apertures, concave valves and strong terminal setae diverging c. 90˚, and diverging from the apical plane (Brunel Group II) with spirally arranged puncta and indistinct transverse striations (Fig 20B and 20C) [18]. This agrees with our observations, except for the spiral arrangement of puncta. Cleve’s drawings (loc. cit. Fig 1C) show a pattern of spines or puncta on the spines similar to what we have seen (Fig 14A–14D), perhaps spirally inserted spines.

Chaetoceros mitra can distinguished from all the other species of the complex by the sibling setae diverging 30˚–80˚ from the apical plane, defined as Brunel Group II (Fig 13A and 13B) (Brunel 1972), as depicted by Cleve ([18], Fig 20C in the present paper). In the other species, setae are positioned more or less in the apical plane, i.e., Brunel Group I (Figs 1C–1D, 4A–4C, 8A–8B and 11A). Hasle & Syvertsen [6] stated that C. mitra, when seen in broad girdle view, has parallel or convergent terminal setae and in this way differs from C. lorenzianus, whose terminal setae are diverging. This does not agree with our findings, the terminal setae in our material of C. mitra were diverging, as illustrated by Cleve [18], Fig 20C.

The morphology of the resting spore as a distinguishing character for C. mitra needs evaluation as similar resting spores can be found in C. elegans (Fig 7A–7F) and were reported in material identified as C. lorenzianus [6, 9, 10, 39]. The differences with the C. elegans resting spore is discussed below and differentiation with the spore of C. lorenzianus must await “true” C. lorenzianus (from the type locality) being brought into culture. The illustration of a resting spore ([11], pl. 38, fig e) appears to disagree with C. mitra due to the length of the elevations, the relationship between the length of the elevation and the processes as well as the slope of the elevation. In an account of fossil Chaetoceros spores by [24, 40], three ‘species’ (form-species) of Dicladia are treated. Suto’s Dicladia mitra is different from our Fig 15B of Chaetoceros mitra . The relationship between the recent and the fossil taxa is unknown.

The size and density of the seta poroids (Fig 14B–14E) differentiates C. mitra from all the other species in the complex (Table 1), as C. mitra has the smallest type of seta poroids (Fig 17), measuring 0.2 ± 0.1 um, a size that can hardly be resolved in LM. The shape of the apertures is similar in C. decipiens and C. mitra, and poroid sizes overlap, but the two species Chaetoceros decipiens and C. mitra differ from each other in the density of setae poroids, presence/absence of fusion of the seta bases, the divergence of the intercalary setae from the apical plane (compare Figs 1D and 13B) and the ornamentation of the valves. While C. decipiens has scattered poroids, C. mitra lacks poroids on the valve face (compare Figs 2H and 14H).