1. The visual surface of schizochroal eyes

The visual surface, and thus the visual field of the schizochroal eye is oriented anteriorly, anterior-laterally, and just slightly backward. The palpebral lobe covers the upper side of the eye. In contrast to, for example, many Silurian phacopine trilobites with relatively large but densely packed lenses (Fig. 1a–d), in Devonian phacopine trilobites, each large lens is separated at half a lens-diameter distance from its neighbour horizontally, while the vertical distance of the lenses is smaller (Schoenemann Reference Schoenemann2021, Reference Schoenemann2025). Horizontally, the visual surface is curved, offering a field of view often of ∼140°, while vertically, the visual surface is quite steep, just slightly tilted backwards, allowing the lower lenses protruding (Fig. 1a–d), providing some visual perception of the water-column above the trilobite. Due to the higher vertical resolution, movements in the vertical direction can be perceived better than in the horizontal direction (Fordyce & Cronin Reference Fordyce and Cronin1989, Reference Fordyce and Cronin1993), but the visual field horizontally is much wider than vertically. Intriguingly, in many phacopines the lenses protrude quite markedly from the visual surface, with the consequence that each lens has an almost hemispherical surface and a wide range (about 180°) to capture light (Fig. 1b), and so these interspaces may be covered optically (Schoenemann Reference Schoenemann2007). Clarkson & Levi-Setti (Reference Clarkson and Levi-Setti1975) showed that the large lenses of phacopine trilobites, in fact, are lens systems. The subunits of the schizochroal eye lenses have different refractive indices, resulting in a complex function of the optical system as a whole. This was suggested to be a system to correct spherical and chromatic aberration, typical of all thick lenses (Clarkson & Levi-Setti Reference Clarkson and Levi-Setti1975). It has been shown more recently that the schizochroal eye actually is a highly specialized system – it is a hyper-compound eye, where below each of the large lenses a small individual compound eye is located (Schoenemann et al. Reference Schoenemann, Clarkson, Bartels, Südkamp, Rössner and Ryck2021)

Figure 1

Aspect and field of view of the typical phacopine compound eye and the injured compound eye of Achatella sp. (a) Drawing of a typical phacopine compound eye (in Acaste downingiae (Salter)), lateral view (Clarkson Reference Clarkson1966, Fig. 1a). (b) Hemispherical lens protruding from the visual surface, gathering light from ∼180°. (c) Drawing of a typical phacopine compound eye (in Acaste downingiae (Salter)) as seen from above (Clarkson Reference Clarkson1966, Fig. 1b). Note that each lens can gather light from above. (d) Visual surface of c seen from the side and from above. (e) Eye with intact visual surface of Achatella sp. RGM 1495535 from the same location as f–m. (f) Injured and healed visual surface of the injured compound eye of Achatella sp. RGM 1495534, The Haerst, Zwolle, NL. (g) Injured specimen with interpretative arrows and lines. Blue line: dent in the visual surface, yellow arrows: restored lenses, smaller than in the normal pattern; black arrows: restored lenses larger than in the normal pattern; red arrow: tiny lens, minuscule but in the correct position. (h) Injured specimen with interpretative markings. Blue: lenses which are smaller; yellow: lenses which are larger than in the normal pattern; green: lenses with irregular shapes; blue line: dent in the visual surface. (i) Small tubercles not just between the lenses, but also above some of the marginal lenses. (j) Example of a part of the intact visual surface. (k) Red rectangle indicates the position of image in i, blue rectangle indicates the position of j. (l) Injured and restored part of the visual surface. (m) Affected region, marked in blue.

The geometry of the schizochroal eye was nicely described by Clarkson (Reference Clarkson1966) who subsequently (Reference Clarkson and Martinsson1975) discussed how it developed during the ontogeny of the trilobite. Starting from the upper side of the eye with a generative zone, during growth the lenses were ordered in a logarithmic spiral, and with every moulting, one row of lenses was added, while the generative zone moved downwards until the final size of the eye was reached. The size of the lenses increased with each moulting. In addition, the spirals grew anteriorly, adding small new lenses. Consequently, a characteristic lens pattern evolved, constant for each species and sometimes even within a family or superfamily (Clarkson Reference Clarkson and Martinsson1975).

While in earlier schizochroal eyes such as in the Silurian Acaste (Fig. 1a–d) the lenses are generally more closely spaced than those in Devonian phacopines, it is remarkable that in the eye of the Ordovician Achatella sp., the more advanced mode is already represented. As Figure 1e shows, Achatella sp. possessed remarkably protruding compound eyes, probably indicating a function of elaborated visual orientation and visually guided behaviour.

2. Analyses of the injured eye of Achatella sp.

The isolated eyes of the Ordovician trilobite Achatella sp. (Fig. 1e–m) were found in a glacial erratic boulder of silicified carbonate from the Upper Sandbian Stage (Haljala/Keila regional stages (CIII–DII)), originating from Northern Scandinavia and the Baltic area (see section 6, Materials and Methods) at The Haerst, near Zwolle, The Netherlands. Although the upper part is missing, the left eye of Achatella sp. in Figure 1e shows the characteristic pattern of a schizochroal compound eye. The visual surface of a second eye (Fig. 1f–m), however, is obviously disturbed. One may argue that the smooth areas within the visual surface are an unprepared matrix covering the visual surface, and the intact lenses lie below. However, there are clearly smaller and larger lenses of irregular size in this area (Fig. 1f–h, l, m), and there is a a conspicuous, linear dent (blue dashed line on Fig. 1g, h). This suggests a healed eye injury, which needs to be described and is interpreted as a healed, restored injury, probably caused by a macrophagous predatory attack. Furthermore, it can provide information on the genetic program of compound-eye formation, at least in phacopine trilobites.

For analysis, each lens in the interpretative graphics was represented by a blue spot of equivalent size, as it was not covered by the matrix, and the visual surface was visible (Fig. 2). The oval shape of lenses in the graphic results from the photograph’s parallax. Then, the centres of the lens representants were connected, indicating the principles of order within the visual surface.

Figure 2

Organisation of the injured and restored eye of Achatella sp. (a) Horizontal rows of lenses, maintained after healing, indicated by the dotted line. (b) Vertical rows of lenses, maintained after healing, indicated by the dotted line. (c) Oblique arrangement of the lenses maintained after healing indicated by the dotted line. (d) All restored lenses (yellow) are still in the correct place (yellow: affected region). (e) Anterior view of the investigated damaged eye of Achatella sp. (RGM 1495534) for comparison.

Figures 1m and 2e show clearly that in the anterior part of the eye, the typical regular lens arrangement pattern, as described above, is still intact, whereas in the central area, it is not, as indicated by the pale blue areas in Figures 2a–c. Several features characterize this destroyed part of the visual surface: 1. There is an indented line across the upper part of the eye (blue dashed line, Fig. 1g, h). 2. Some lenses are smaller than expected by the regular pattern (see anterior part of the eye), indicated by yellow arrows and blue transparent circles (Fig. 1h). 3. Some lenses are larger than expected (indicated by dark green arrows (Fig. 1g) and yellow transparent circles (Fig. 1h). 4. The general pattern of vertical and horizontal order of the lenses is retained (Fig. 3a–e), even if a substituted lens is very small (red arrow, Fig. 1g), 5. Some of the lenses adjacent to the indented line are not round but of irregular shape. (Fig. 1h in green).

There are lines of tubercles of unknown function within the cuticular interspaces between the lenses (Fig. 1i). Remarkably, the same tubercles occur in a few lenses adjacent to the smooth areas. If these tubercles are built on the lenses and they possibly were of similar character as those within the interspaces, the lenses may not have been transparent or fully functional.

3. Healed injuries in trilobites

Healed trilobite injuries are frequently reported, mainly caused by macrophagous predatory attacks on the trilobite body (for an overview, see, e.g., Owen Reference Owen1985; Rudkin Reference Rudkin1985; Babcock Reference Babcock1993; Bicknell & Pates Reference Bicknell and Pates2020; Bicknell et al. Reference Bicknell, Smith, Howells and Foster2022). They must be distinguished from other malformations, such as those caused by teratological reasons, damage during moulting or other activities and also from characteristic swellings, often indicating parasitic violations, genetic malformations or the misformation of individual lenses in the compound eyes (e.g., in Owen Reference Owen1985, Bicknell et al. Reference Bicknell, Smith and Hopkins2024, Reference Bicknell, Smith, Amati and Hopkins2025).

Healed injuries are mainly reported from the lateral areas of the trilobite thorax and pygidium, more rarely from the cephalon or the rachis, indicating that the latter were more often lethal than the former (Owen Reference Owen1985). This is understandable if we assume that the heart of trilobites, as typical for all tetraconates, was an elongated tube and lay dorsally along the rachis. The open circulatory system of these animals inevitably leads to bleeding to death in the event of an injury, especially if this is located near the heart, and if the immune system does not act quickly enough closing the wound. While evidence of injuries of the cephalon are rare, a good example of a complicated laceration of an eye and its healing was described for the trilobite Telephina intermedia (Thorslund Reference Thorslund1935) by Schoenemann et al. (Reference Schoenemann, Clarkson and Høyberget2017b). In this specimen, the indented area inside the visual surface (Schoenemann et al. Reference Schoenemann, Clarkson and Høyberget2017b, Fig. 1e, f) and the imperfect restoration of obviously lost lenses into the now smooth and structureless area of the eye surface indicate an incompletely healed destruction of the visual surface. Other reports of injured cephala were provided by Babcock (Reference Babcock1993), Fatka et al. (Reference Fatka, Budil and Grigar2015), Pates & Bicknell (Reference Pates and Bicknell2019) and Bicknell et al. (Reference Bicknell, Smith and Hopkins2024).

4. Interpretation of the injured eye

Healing processes, as regarded here and by other authors, reflect the existence of functioning immune and repair mechanisms, which probably evolved before or during the early Cambrian (e.g., Morris & Jenkins Reference Morris and Jenkins1985; Jago & Haines Reference Jago and Haines2002; Fatka et al. Reference Fatka, Budil and Grigar2015; Pates et al. Reference Pates, Bicknell, Daley and Zamora2017; Bicknell & Pates Reference Bicknell and Pates2020; Bicknell et al. Reference Bicknell, Smith, Howells and Foster2022). Zong (Reference Zong2021) also described the difficulties and dangers that may arise from exoskeleton irregularities due to the healing processes. Since, in the case of the eye of Achatella sp., the visual surface is closed, and there is a complete, albeit altered lens pattern, the animal to which this eye belonged, must already have finished at least one moult after the lesion. It is an interesting question, whether the lesion occurred during a moult, which made it ‘easier’ to be repaired during an ongoing process of forming an eye, or whether it was possible to repair the leak during ‘normal’ life of the trilobite, in the time between two moults. Unfortunately, we probably cannot solve this question.

The trilobite survived the injury, and the way the repair was carried out provides information about the genetic program that determines the formation of the pattern of a visual surface in a schizochroal compound eye. We interpret the linear furrow as a line of indenting and tearing of the visual surface. Below that, due to a greater injury, lies a smooth lensless area, framed in its upper part by some lenses of irregular size. Inside the smooth area, most original lenses are missing, just some new lenses were installed, and intriguingly, the vertical and horizontal order of the new lenses is retained, even after the relatively great destruction of the visual surface. The size of the restored lenses is often smaller or larger than originally. There are smaller lenses at the lower margin of the injured zone, indicating where the injury finishes.

The entire lesion has a dimension of about 750 × 750 µm. The irregular shape and the wide flat area filled with homogeneous replacement tissue suggest a hole closed by a successful healing process. Such a hole could have been caused by a sharp ‘tool’, such as the sharp beak of a small cephalopod, as has been suggested in different trilobites (Schoenemann et al. Reference Schoenemann, Clarkson and Høyberget2017b, Fatka et al. Reference Fatka, Budil and Mikuláš2022). Orthocerids are among the most important predators of the Ordovician period and were equipped with a pointed, parrot-like beak, and certainly represent serious predators of trilobites. Beaks of fully grown orthocerid, however, may usually be too large to injure a small trilobite eye, as here, some millimetres in size. Crustaceans of that time, bottom dwellers such as ostracodes or phyllocarids might have left larger bite marks, or had a lifestyle other than predatory (Brahm & Geiger Reference Brahm and Geiger1966; Williams et al. Reference Williams, Perrier, Bennett, Hearing, Stocker and Harvey2015).

An unsuccessful attack by a eurypterid might be considered. Eurypterids appeared at the beginning of the Ordovician (Wang et al. Reference Wang, Braddy, Botting and Zhang2023) but so far it seems that the Baltic Ordovician is devoid of eurypterids; they appear there in the Silurian in the fossil record (Tetlie Reference Tetlie2007). Perhaps, however, the injured trilobite eye is a first sign of their existence during the Baltic Ordovician. The scale of the injury would be compatible with the fact that among the earliest eurypterids were some smaller forms, such as Orcanopterus manitoulinensis Stott et al. 2005, Brachyopterus stubblefieldi Størmer 1951, Onychopterella augusti Braddy, Aldridge & Theron 1995, and Paraeurypterus anatoliensis Lamsdell, Hosgör & Selden Reference Lamsdell, Hoşgör and Selden2013 (P. Selden pers. communication 2025). An abrasion of the visual surface by feeding from conspecifics shortly after moulting in a soft-shelled stage is rather unlikely due to the linear dent in the visual surface. Thus, the origin of the lesion remains unresolved.

Finally, considering these findings, we can find a hierarchic genetic program for establishing an ordered visual surface, at least for the schizochroal eyes of trilobites.

The first superordinate instance is the vertical and diagonal arrangement of the lenses, as established by the mode of growing. It is retained/repeated during growth, while at each new moulting, a new row of lenses and anterior lenses are added. The pattern continues during each moult and is retained and enforced during the restoration of an extended injury. It must be connected with each genetic command to establish a lens. The second instance is the size of the lenses. The appropriate size, as expected by the regularity of the lens pattern within the visual surface, is not always retained. Some of the lenses, especially along the tearing line and the outer margin of the lesion, are too small – probably because of a non-intact generative equipment, and the restoration probably depends on the possibilities and conditions of the restoration site. Significantly, some of the lenses restored in the more central part of the now smooth, injured area are larger than would be expected by order of the original pattern, and may indicate that the system has some capacity for regulation regarding the light-gathering capacity (large lenses with a higher light gathering capacity ‘trying’ to compensate for the loss of the other lenses).