Type species.—Fuxianospira gyrata Chen and Zhou, 1997, from the Yuanshan Formation, Chengjiang, Yunnan, China, by monotypy.

2021 Fuxianospira gyrata Chen and Zhou, 1997; Wang et al., 2021, p. 3, figs. 4–6 (includes a synonymy).

Holotype.—Early Life Research Center, Chengjiang ELRC80001, Yuanshan Formation, Chengjiang, Yunnan, China (Chen and Zhou, 1997, p. 88, fig. 143).

Description.—Masses of fragments of poorly preserved, carbonaceous strap-like features are preserved as reflective films (Fig. 2.1) or, in some cases, with a covering of iron oxide, presumably the result of weathering pyrite (Fig. 2.2). The straps are mainly straight, but many show curvature and folding, and they occasionally appear to branch. They are parallel sided, ∼0.5–1.0 mm wide, with irregular undulating margins and convex terminations. They rarely exceed 2 cm in length. No structure is evident on the cuticle.

Material.—YPM IP 239095, part and counterpart; YPM IP 239097.

Discussion.—The specimens fall into Morphogroup 1 of LoDuca et al. (2017) among early Paleozoic non-calcified macroalgae, and their morphology is consistent with that of Fuxianospira, which is common in the Chengjiang biota (LoDuca et al., 2015b, 2017; Wang et al., 2021). The undulating margins reflect the coiled structure, although the preservation, like that of occurrences of Fuxianospora elsewhere (LoDuca et al., 2015b; Wang et al., 2021), is inadequate to show the morphology in detail. The apparent branching is likely the result of overlap (see LoDuca et al., 2015b, fig. 1). The thalli of Sinocylindra, which is also from the Chengjiang biota, tend to be longer and thinner, with smoother margins (Wang et al., 2021). We follow LoDuca et al. (2015b) and Wang et al. (2021) in tentatively assigning Fuxianospira to Chlorophyta.

Remarks.—Similar specimens from other Cambrian localities were assigned mainly to Yuknessia simplex Walcott, 1919, which was originally described from the Burgess Shale and interpreted by Walcott (1919) as an alga. LoDuca et al. (2015a) showed that most of Walcott's Burgess Shale specimens preserve banding on the cuticle and identified them as a benthic colonial pterobranch hemichordate (see also Ramírez-Guerrero and Cameron, 2021); just a few of Walcott's specimens turned out to be algal. However, LoDuca et al. (2015b) also investigated specimens attributed to Yuknessia from the Chengjiang biota in China and other Cambrian Lagerstätten in North America, and assigned them to the alga Fuxianospora Chen and Zhou, 1997.

Type species.—Leptomitus zitteli Walcott, 1886, Parker Formation, Parker Quarry, Franklin County, Vermont, USA.

1987 Leptomitus zitteli Walcott, 1886; Rigby, p. 453, fig. 3.3–3.5 (includes a synonymy).

2021 Leptomitus; Pari et al., p. 693, fig. 2I.

Lectotype.—USNM 15308, Parker Formation, Parker Quarry, Franklin County, Vermont, USA. Cambrian Series 2, Stage 4 (Rigby, 1987, fig. 3.3–3.5).

Occurrence.—Leptomitus zitteli is presently known only from the Parker Formation at Parker's Cobble (Rigby, 1987; García-Bellido et al., 2007).

Description.—Rigby (1987) described the spicular structure of this elongate tubular sponge, which comprises an outer layer of parallel oxeas and an inner layer of tiny horizontal spicules (Fig. 3.1, 3.3). The lectotype (USNM 15308) reveals a plumose arrangement of spicules at the distal end (Fig. 3.2), which may represent their arrangement in the interior of the skeleton (Rigby, 1987). It is difficult to distinguish between coarse and fine long vertical oxeas and to determine their length. Dimensions of the sponge vary between specimens. YPM IP 543326 is a fragment ∼70 mm long consisting of a bundle of oxeas, roughly aligned, with some diverging at either end (Fig. 3.4). The specimen is preserved in pyrite, which prevents individual spicules from being distinguished. USNM 15308, which is ∼70 mm long, varies in width from 10–2 mm along its length. YPM IP 239151 (Fig. 3.5–3.7) is longer than other specimens from this locality at 12 cm, although incompletely exposed, but only 2–4 mm wide along its length. The distribution of the spicules indicates some decay and exfoliation along its length with small fragments displaced laterally and evidence that the outer layer has separated at one point about halfway along the preserved length, giving the appearance of branching. This is consistent with the occurrence of sponge fragments such as YPM IP 543326 (Fig. 3.4) and YPM IP 239096.

Material.—USNM 15308, lectotype, and USNM 419730 and 419732 on the same block, part and counterpart. USNM 419731, part and counterpart, on a different block, YPM IP 239096, YPM IP 239151, YPM IP 239152, YPM IP 543326.

Remarks.—Rigby (1986) published a generalized reconstruction of Leptomitus based on his revision of L. lineatus (Walcott, 1920) from the Burgess Shale. The following year he provided a full description of Walcott's material of L. zitteli, and a synonymy (Rigby, 1987). He designated as lectotype the best-preserved specimen, USNM 15308, part and counterpart, which Walcott (1886, legend to plate 2) referred to as the type specimen (the other two specimens on the same block now carry the numbers USNM 419730 and 419732). A specimen on a second block and two non-contiguous fragments of its counterpart were also originally included under USNM 15308, but that individual is now numbered USNM 419731.

Type species.—Protospongia fenestrata Salter, 1864, Menevian Group, Porthyrhaw, St. Davids, Wales.

1987 ?Protospongia hicksi Hinde, 1887; Rigby, p. 459, fig. 4.

Holotype.—Sedgwick Museum, Cambridge SM A1035, Menevian Group, Porthyrhaw, St. Davids, Wales.

Occurrence.—In addition to this probable occurrence in the Parker Formation at Parker's Cobble (Rigby, 1987), Protospongia hicksi is present in younger North American Cambrian Lagerstätten, including the Burgess Shale (Rigby, 1986) and the Trilobite Beds on Mount Stephen (Rigby and Collins, 2004). Spicules closely resembling those of Protospongia also have been reported from the upper part of the Kinzers Formation near Lancaster, Pennsylvania, which is similar in age to the Mount Stephen Trilobite Beds (Campbell, 1971).

Material.—Fragments and isolated spicules associated with USNM 15308, USNM 419731, and USNM 419732.

Remarks.—We follow Rigby (1987) in tentatively assigning these spicules to ?Protospongia hicksi. In doing so, however, we acknowledge that their morphology is poorly constrained. Protospongia and similar taxa are in need of revision (Botting and Muir, 2018) and may be more appropriately assigned to Ascospongiae than to Hexactinellida (Botting, 2021). Rigby (1987) recorded ?Protospongia hicksi in intimate association with the specimens of L. zitteli on the slab that includes the lectotype USNM 15308. Rigby (1987, p. 459) noted that the lectotype, in its middle part, appears to have “buried a partially intact fragment of a protosponge” and (1987, p. 460) “isolated Protospongia-like stauracts occur scattered throughout matrix between Leptomitus fragments.” He also observed first order quadrules associated with USNM 419732. These Protospongia spicules were not noted by Walcott (1886). We illustrate examples of these cruciform spicules from Walcott's original specimens here (Fig. 3.8, 3.9). We did not find examples associated with the specimens of Leptomitus discovered during our excavations.

Type species.—Ottoia prolifica Walcott, 1911 (Walcott, 1911b), Burgess Shale, British Columbia, Canada.

2021 Priapulid reminiscent of Ottoia; Pari et al., p. 695, fig. 2G.

Description.—The specimen, YPM IP 239053, which is ∼9.5 cm long, with a maximum width of 8.5–9 mm, is tightly coiled (Fig. 4). It narrows gradually toward one end, curving sharply and expanding slightly at the termination, resulting in an outline reminiscent of a proboscis. The other end tapers more abruptly. Details are poorly preserved, but the body preserves evidence of transverse annulations and a wide reflective structure in the anterior portion that may represent the gut.

Material.—YPM IP 239053.

Remarks.—The annulations and presumed proboscis suggest a priapulid affinity, and the outline of the specimen is similar to Ottoia, among Cambrian genera. Smith et al. (2015) demonstrated the value of sclerites in distinguishing species of ottoiids and they cautioned against identifying indeterminate priapulids that do not preserve such characters. On this basis, we assign our specimen to Ottoia with a query. Smith et al. (2015, p. 706) noted that “there are no confirmed records of Ottoia macrofossils outside the Burgess Shale,” but Y. Yang et al. (2016) described a new species based on multiple body fossils from the Kaili Formation in South China, demonstrating the occurrence of the genus outside Laurentia. Resser and Howell (1938, p. 215) tentatively assigned a specimen from the Kinzers Formation to Ottoia, but Conway Morris (1977, p. 85) identified it as a palaeoscolecid. If the Parker Formation specimen were Ottoia, it would be the oldest macrofossil representative of the genus discovered to date. Sansom (2016) noted the decay-resistant nature of sclerites in living priapulids, and Howard et al. (2020) indicated that their absence is a feature of stem ecdysozoans. YPM IP 239053 preserves cuticle annulations in places, but the distribution of carbon on the specimen and the associated Vermontcaris montcalmi n. gen. n. sp., is patchy (fig. 1), presumably as a result of degradation. Thus the apparent absence of scalids on the presumed proboscis may or may not be real (Howard et al., 2020). Thus the fossil may represent a stem ecdysozoan similar to Acosmia maotiania Chen and Zhou, 1997, rather than a stem priapulid.

The specimen shows some similarity to Atalotaenia adela García-Bellido and Conway Morris, 1999, from the Kinzers Formation (García-Bellido and Conway Morris, 1999). However, the single reported specimen of A. adela is incomplete, it preserves a ribbon-like structure for which there is no equivalent in YPM IP 239053, and the annulations are more widely separated. García-Bellido and Conway Morris (1999) assigned A. adela to Phylum, Class, Family uncertain.

2021 Radiodont; Pari et al., p. 695, fig. 2F.

Description.—The specimen, which is preserved in lateral view, is comprised of the distalmost podomere of the shaft, which bears a large endite (= ventral spine: see Guo et al., 2019), and an articulated region of 13 podomeres, each bearing ventrally projecting endites with short lateral auxiliary spines (Fig. 5). The proximal part of the shaft and the endites, particularly in the more distal part of the appendage, are incomplete, which limits the detail available. Decay and disarticulation have resulted in separation of the appendage into two portions at the junction between podomeres 5 and 6 in the distal articulated region; the junction between podomeres 1 and 2 is also partially ruptured. Podomeres 6–9 of the articulated region are a little longer than those proximal to them, and the height (measured from the mid-point of the dorsal margin to apparent attachment of the endite) to length ratio of the podomeres averages ∼2.0. It is not clear whether the endites are paired or not. The podomeres are separated by a gap, tapering dorsally, which represents the position of the articulating membrane. In spite of poor preservation it is clear that the endite on the distalmost podomere of the shaft and on the first podomere of the articulated region are enlarged (hypertrophied) compared to the rest (Fig. 5). The endite on podomere 1 of the articulated region is the largest, and three auxiliary spines radiate from its distal margin. The endite on podomere 2 is shorter than that on podomere 3. The endite on podomere 5 appears enlarged and was presumably longer than on podomere 3, based on the width at its base. Long robust dorsal spines are present on the distalmost podomeres, which curve strongly ventrally, and the thirteenth podomere bears two terminal spines. The specimen measures ∼140 mm along its dorsal margin excluding the dorsal spines on the distal segments and the gap resulting from disarticulation.

Material.—YPM IP 239052, part and counterpart.

Discussion.—This specimen represents the first record of a radiodont from the Parker Formation and adds a new locality for this group on the Laurentian paleocontinent (Pates and Daley, 2019).

Remarks.—Although radiodont families are diagnosed on the basis of the complete morphology of the animal, and the mouth parts are particularly characteristic (Cong et al., 2018), specimens are often assigned, even to species, based on the frontal appendages (Pates et al., 2021). As far as can be determined, the morphology of the Parker Formation appendage does not correspond precisely to that of any previously described radiodont (see Cong et al., 2018; Guo et al., 2019; Pates and Daley, 2019; Pates et al., 2021). It is incomplete proximally but likely comprises the distalmost podomere of the shaft, which bears a large hypertrophied endite, and an articulated region of 13 podomeres. It shows greatest similarity to Anomalocarididae and Amplectobeluidae, both of which include genera with paired endites. YPM IP 239052 differs from Anomalocarididae in the relative lengths of the endites and details of the auxiliary spines (Guo et al., 2019; Pates and Daley, 2019; Pates et al., 2021). It shows greater similarity to the frontal appendage of Amplectobeluidae, which has a single hypertrophied proximal endite on the first podomere of the articulated portion, high/deep podomeres 12 or 13 in number, robust dorsal spines on the distalmost podomeres, and an endite on podomere 5 of the articulated region that is enlarged relative to that on podomere 3 (Cong et al., 2018; Guo et al., 2019; Pates et al., 2021). This last character is particularly evident in Ramskoeldia from the Chengjiang biota of China (Series 2, Stage 3), but that genus lacks a large endite on the shaft (Pates et al., 2021). The large height to length ratio of the podomeres, large first endite on the articulated region, and pronounced distal dorsal spines are shared with Amplectobelua (Daley and Budd, 2010; Lerosey-Aubril et al., 2020), but the articulated region of the Parker Formation appendage includes 13 as opposed to 12 podomeres. The Parker Formation appendage shares more characters with Laminacaris chimera Guo et al., 2019, from the Chengjiang biota than with Amplectobelua, including a large endite on the shaft, 13 podomeres in the articulated region, and a larger endite on podomere 5 of that region than podomere 3 (Guo et al., 2019). Laminacaris has been recovered as an amplectobeluid in some phylogenetic analyses (Lerosey-Aubril and Pates, 2018; but see Moysiuk and Caron, 2021).

The phylogeny of radiodonts is not well resolved, but Anomalocarididae and Amplectobeluidae fall within the same clade (Cong et al., 2014; Vinther et al., 2014; Van Roy et al., 2015; Lerosey-Aubril and Pates, 2018; Moysiuk and Caron, 2021). The available evidence supports assignment of the Parker appendage to Amplectobeluidae. It shows similarities to a number of genera, but we do not assign it beyond family here in view of the incomplete preservation. The Parker Formation appendage may represent a new genus and species within this group.

Type species.—Tuzoia retifera Walcott, 1912 (Walcott, 1912a), Burgess Shale, British Columbia, Canada.

1890 Problematical organic marking; Walcott, pl. 59, fig. 1.

1938 Tuzoia vermontensis Resser and Howell, p. 231, pl. 13, fig. 1.

2003 Tuzoia polleni Resser, 1929; Lieberman, p. 681.

2007 Tuzoia polleni Resser, 1929; Vannier et al., p. 460, fig. 17.2, p. 462 (includes a synonymy).

2021 Tuzoia polleni Resser, 1929; Pari et al., p. 694, fig. 2B, C.

Holotype.—USNM 80485, Eager Formation, Cranbrook, British Columbia, Canada (Resser, 1929, pl. 5, fig. 1).

Description.—USNM 26728 (Fig. 6.1, 6.2) represents a single right valve with an unusual subcircular outline. It preserves four spines on the posterior margin and perhaps five along the hinge line (only the posteriormost of these five is clearly preserved, the others are somewhat equivocal and have been outlined in pencil on the specimen). YPM IP 239051 (Fig. 6.3, 6.4) and YPM IP 239150 (Fig. 6.5, 6.6) also reveal the complete valve outline. The ventral margin of most of the specimens is followed by a well-delineated wide margin (Fig. 6.3–6.8), which may be modified by compression. YPM IP 239051 (Fig. 6.3, 6.4) preserves at least four spines around the posterior margin. There is also at least one spine on the anterior margin and there may be several on the hinge line, but these cannot be discerned with confidence due to the nature of the preservation. YPM IP 239150 (Fig. 6.5, 6.6), which preserves both valves slightly offset, preserves two slender posterior spines on each valve. YPM IP 239149 (Fig. 6.7, 6.8) and YPM IP 543331 (Fig. 6.9, 6.10) are incompletely exposed. They preserve four and seven spines, respectively, around the posterior margin; in neither case are spines evident on the hinge. Most of the specimens preserve a trace of the lateral ridge (Fig. 6.1–6.4, 6.7, 6.8) roughly aligned with the mid-posterior spine, its position and shape varying with the orientation and flattening of the carapace. Spines are not evident along the ridge, but these are difficult to discern when flattened onto the valve.

USNM 26728 (Fig. 6.1, 6.2) is 8 cm long and 7.6 cm high excluding spines (height/length, H/L 0.95). YPM IP 239051 (Fig. 6.3, 6.4) is 7.4 cm long and 4.7 cm high (H/L 0.64). YPM IP 239149 (Fig. 6.7, 6.8) and YPM IP 543331 (Fig. 6.9, 6.10) are ∼3.9 cm high and ∼6.2 cm high, respectively, based on extrapolations from the exposed parts of the valves. YPM IP 239150 (Fig. 6.5, 6.6) preserves both valves offset; the completely exposed (?left) valve is ∼4.8 cm long and 3.2 cm high (H/L 0.67).

Material.—USNM 26728, the only previously known specimen of Tuzoia from this locality, and four new specimens with part and counterpart: YPM IP 543331, YPM IP 239051, YPM IP 239149, YPM IP 239150.

Discussion.—The wide ventral margin in the new specimens from the Parker Quarry Lagerstätte is similar to that in Tuzoia sp. from the Emu Bay Shale of Australia, which was interpreted as a doublure (García-Bellido et al., 2009). Variability in the preservation of the carapace outline and spines makes the identification of species of Tuzoia difficult (Vannier et al., 2007). The shape and number of spines may change during ontogeny (Wen et al., 2015). The number of spines evident around the margin of the valves also shows intraspecific variability, which may reflect morphologic differences and/or taphonomic processes (Liebermann, 2003; Vannier et al., 2007). USNM 26728 (Fig. 6.1, 6.2) shows a much greater height to length ratio than YPM IP 239051 (Fig. 6.3, 6.4) and YPM IP 239150 (Fig. 6.5, 6.6). This might be a result of burial of USNM 26728 with the valve tilted at an angle to bedding and consequent reduction of the length, but, apart from a hint of concentric lines close to the anterior margin of the valve, there is no independent evidence that this is the case. Specimens of the new arthropod Vermontcaris montcalmi n. gen. n. sp., however, clearly vary in outline as a consequence of flattening in different attitudes to bedding following transport and burial (Figs. 7–9). Such variation in orientation is analogous with that in the Burgess Shale and is consistent with evidence that the Parker Quarry Lagerstätte sediments were deposited from dense mud-rich flows (Pari et al., 2021). Carapaces of Tuzoia from other localities are also characterized by variation in outline (Vannier et al., 2007).

If the length of USNM 26728 (Fig. 6.1, 6.2) does reflect compaction at an angle to bedding, its original length can be estimated based on the height to length ratio of YPM IP 239051 (Fig. 6.3, 6.4), which would extend it from 8 cm to 11.9 cm. This exceeds the maximum length of ∼10 cm reported for T. polleni by Vannier et al. (2007, p. 462). Valves of Tuzoia up to ∼18 cm long have been reported from the Cambrian Miaolingian (Drumian) of Bohemia (Chlupác and Kordule, 2002, fig. 3), but the largest North American examples previously noted, specimens of T. retifera from the Burgess Shale, reached maximum lengths of ∼12 cm (Vannier et al., 2007).

Remarks.—Walcott (1890) figured USNM 26728 (Fig. 6.1, 6.2), without description, noting it as “a problematical organic marking,” and the specimen was subsequently labeled as Medusa? sp. (Resser and Howell, 1938, p. 231). Resser and Howell (1938) described it as Tuzoia vermontensis, a new species of the genus that Walcott (1912a) had erected for Burgess Shale material. Lieberman (2003) synonymized all the species of Tuzoia (T. polleni, T. nodosa Resser, 1929, and T. spinosa Resser, 1929) that Resser (1929) originally described from the Eager Formation, assigning them to T. polleni. Lieberman (2003) did not examine the original specimen of T. vermontensis, USNM 26728, but interpreted it as a single valve and synonymized it with T. polleni because of the presence of spines on the entire margin. Vannier et al. (2007) went further and synonymized T. nitida Resser and Howell, 1938, from the Kinzers Formation with T. polleni. They tentatively retained T. vermontensis as a synonym of T. polleni (Vannier et al., 2007, p. 463), but they interpreted the spines along the dorsal margin (hinge line) as ridge spines on the right valve (Vannier et al., 2007, fig. 17.2) based on their erroneous interpretation of USNM 26728 as a dorsoventrally compacted (‘butterflied’) specimen showing both valves.

The Parker Quarry Lagerstätte specimens vary in outline and the distribution of spines (Fig. 6). Identification is compromised by the lack of a dorsoventrally flattened specimen, which might show spines on the lateral ridge; clear evidence of reticulation is also absent. The morphological features of the Parker Quarry Lagerstätte specimens show a general similarity to T. canadensis Resser, 1929, as well as to T. polleni (Pari et al., 2021; see Vannier et al., 2007, fig. 25.3, 25.5). We follow Lieberman (2003) and Vannier et al. (2007) in synonymizing the Parker Formation T. vermontensis with T. polleni. We note, however, that future discoveries and analysis may show otherwise, or even that the new specimens may represent a different species to Walcott's original, in which case more than one species is present in the Parker Formation, as in the Burgess Shale (Vannier et al., 2007) and the Balang Formation of Guizhou, China (Wen et al., 2019).

Type species.—Protocaris marshi Walcott, 1884, Parker Formation, Parker Quarry, Franklin County, Vermont, USA.

1976 Protocaris marshi Walcott, 1884; Briggs, p. 2, text figs. 1, 3, pl. 1, figs. 1–3 (includes a synonymy).

2008 Protocaris marshi Walcott, 1884; Budd, p. 567.

2021 Protocaris marshi Walcott, 1884; Pari et al., p. 693, fig. 2A.

Holotype and only known specimen.—USNM 15400, Parker Formation, Parker Quarry, Franklin County, Vermont, USA.

Discussion.—In the absence of additional specimens we cannot add to the description by Briggs (1976). Protocaris marshi is arguably the first soft-bodied Cambrian arthropod described in the literature (Walcott, 1884; see Pari et al., 2021). Simonetta and Delle Cave (1975) assigned P. marshi to a new Order Protocaridida, without a diagnosis. Budd (2008) noted the likely presence in P. marshi of an ‘anterior sclerite’ based on the illustrations in Briggs (1976), a feature that Budd identified in a diversity of early arthropods, but otherwise P. marshi has received relatively little attention. Legg et al. (2013, fig. 4b) placed it on the stem leading to Euarthropoda. Aria and Caron (2017) amended Protocarididae to include Branchiocaris, Tokummia, and Loricicaris, noting that the assignment of Protocaris to their newly diagnosed Protocarididae is provisional. Aria and Caron’s (2017) new diagnosis of Protocarididae includes multiple appendage characters, which are not preserved in Protocaris. They assigned Protocarididae to an emended Order Hymenocarina and recovered them as stem mandibulates rather than on the stem of euarthropods.

Remarks.—USNM 15400 (Fig. 6.11; see Briggs, 1976, text fig. 3, pl. 1, figs. 1–3), the holotype and only known specimen, was redescribed by Briggs (1976) who provided a full synonymy. A specimen from the Kinzers Formation of Pennsylvania (USNM 90826, counterpart in the North Museum, Franklin and Marshall College, P-A-392) referred to this taxon by Resser and Howell (1938), was rejected as unidentifiable (Briggs, 1976).

Type species.—Vermontcaris montcalmi n. gen. n. sp.

Diagnosis.—As for species.

Etymology.—After the State of Vermont in which the locality occurs, plus Latin caris for crab.

Remarks.—Walcott (1886) compared the specimens of this taxon available to him to graptolites, referring them to Diplograptus? simplex (see Conway Morris, 1993, for a history of early research). Resser and Howell (1938) assigned the material to a new genus, Emmonsaspis, noting (p. 233) that “the central rod, the ribbing, and the general shape of this animal argue strongly for its reference to the chordates.” Shaw (1955), in contrast, regarded the affinities of the animal as unknown. Conway Morris (1989, 1993) noted that most of the specimens are similar to the Burgess Shale bivalved arthropod Perspicaris Briggs, 1977. He realized that the central rod, presumably interpreted by Resser and Howell (1938) as a notochord, is the gut and he also observed evidence of a carapace and trunk, and occasional traces of appendages (Conway Morris, 1993). All but one of the specimens figured by Resser and Howell (1938, pl. 9, figs. 2–4) are examples of this arthropod (Fig. 7.1–7.6), which we assign, together with new material (Figs. 9, 10), to Vermontcaris n. gen. because they differ from previously described taxa. The other figured specimen (USNM 15314A: Resser and Howell, 1938, pl. 9, fig. 1) is the lectotype of the chordate Emmonsaspis cambrensis (Walcott, 1890) (Conway Morris, 1993; Conway Morris and Caron, 2014) (see Remarks on Emmonsaspis cambrensis).

1886 Diplograptus? simplex Walcott, part, p. 92, pl. 11, fig. 4 (non fig. 4a).

1889 Phyllograptus? simplex Walcott, part, p. 388.

1890 Phyllograptus? cambrensis (Walcott), part, p. 604, pl. 59, fig. 3a (non fig. 3).

1938 Emmonsaspis cambriensis (Walcott); Resser and Howell, part, p. 233, pl. 9, figs. 2–4 (non fig. 1).

1989 Arthropod, ?comparable to Perspicaris; Conway Morris, p. 278.

1993 Arthropod, ?comparable to Perspicaris; Conway Morris, p. 600.

2021 New bivalved arthropod; Pari et al., p. 693, fig. 2D, G.

Holotype.—YPM IP 543320.

Diagnosis.—Carapace with hinge line, valves suboval, eyes pedunculate, borne on a projection of the cephalon, trunk divided into a thorax of up to 40 short segments, similar in length, bearing elongate biramous appendages, and an apodous abdomen of 4 or 5 longer segments increasing in length posteriorly. Short telson tapering to a point, flanked by leaf-like appendages borne by the posteriormost abdominal somite.

Description.—The head region is concealed by the carapace. Only a projection bearing the eyes extends beyond the valves. The eyes are borne on stalks attached a short distance from the end of this projection (Figs. 8.3, 8.4, 9.7, 9.8). They are suboval in outline, whether flattened parallel to bedding (Fig. 8.3, 8.4, where they are rotated relative to the carapace) or in lateral aspect (Figs. 8.5, 8.6, 8.8–8.10, 9.9, 9.10), indicating that they were ovoid in shape. None of the specimens preserves evidence of antenniform limbs. Small, poorly defined paired structures, one positive, the other negative in relief, are evident in the anterior part of the carapace of some specimens (Figs. 8.8–8.10, 9.1, 9.2); they may represent the impression of a more heavily cuticularized appendage, perhaps equipped for feeding. Specimens in which the carapace is flipped forward (Figs. 8.5, 8.6, 8.8–8.10, 9.5–9.10) indicate that it was only attached in the head region; no evidence of adductor muscle scars is preserved. The valves are elongate oval, but the preserved outline varies with attitude to bedding (compare Fig. 9.1, 9.2 with 9.3, 9.4).

The bivalved carapace covers most of the trunk, which is divided into two tagmata referred to here as thorax and abdomen. Only the posteriormost segments of the thorax and the abdomen extend beyond the valves (Fig. 8.3, 8.4), except where the carapace is flipped forward. Where the anteriormost part of the thorax is exposed (Figs. 8.8–8.10, 9.7, 9.8) it appears poorly preserved, suggesting that the cuticle normally protected by the carapace was lightly sclerotized and more prone to decay. The thoracic segments bear appendages, whereas the abdomen is apodous. The nature of the preservation does not reveal details of the thoracic appendages, but there is evidence of an elongate branch that tapers distally, terminating in a point. Podomere boundaries are not preserved. This branch, presumably the endopod, extends beyond the continuous outline of the appendages (Figs. 8.5, 8.6, 8.8–8.10, 9.5, 9.6), which probably indicates the extent of overlapping flap-like exopods. Thus, although the detailed nature of the appendages is unknown, they appear to be biramous.

The trunk segment boundaries are fringed by spines, which are occasionally preserved in outline at the trunk margin (Figs. 8.1–8.4, 9.3, 9.4). The boundaries between thoracic segments are usually poorly preserved, particularly at the anterior of the trunk. Where they are evident (Figs. 8.1, 8.2, 9.3–9.8), the posterior thoracic segments appear to be roughly equal in length. The number of segments can be estimated by assuming that there was one pair of appendages per segment, although there may have been more, as in Branchiocaris pretiosa (Resser, 1929) (Briggs, 1976). Small relief structures corresponding to thoracic segments (Figs. 8.1, 8.2, 9.5, 9.6) may represent the attachment sites of appendages. The outline of the appendages is never clear, but extrapolation based on their apparent outline in YPM IP 543316 (Fig. 8.3, 8.4) indicates >30, and based on trunk segment divisions, limbs, and attachment sites in YPM IP 239148 (Fig. 9.5, 9.6), >37.

The segments of the abdomen differ from those of the thorax in their greater length and lack of limbs. Identifying the boundary between thorax and abdomen is difficult because there is no abrupt change in the length of segments and the limbs often project posteriorly, overlapping the abdomen. There are at least four, probably five, abdominal segments (Figs. 8.1–8.4, 9.5, 9.6).

The telson is poorly preserved and details are often obscured by superimposition of associated appendages as a result of flattening of the specimens in lateral aspect. A triangular dorsal projection (Fig. 8.3, 8.4), which represents the telson, appears to bear some spines ventrally (Fig. 8.1, 8.2). The anus lies ventral of this projection in the more proximal part of the telson (Fig. 9.7, 9.8). A pair of leaf-like appendages, fringed with fine spines (Fig. 8.1–8.4), appears to articulate with the ventral portion of the pre-telson segment (Fig. 8.1–8.4).

Etymology.—In honor of the Montcalm family on whose property the quarry lies.

Material.—YPM IP 543320 holotype, USNM 15314D–F, USNM 15314 G (three poorly preserved specimens), YPM IP 239148, YPM IP 543312, YPM IP 543313, YPM IP 543316, YPM IP 543318–543322, YPM IP 543324, YPM IP 543325, YPM IP 543333–543335, YPM IP 239153.

Dimensions.—Total length (from the anterior of the carapace to the posterior margin of the telson) is difficult to measure accurately due to the curvature of many specimens. It ranges from ∼26 mm (Figs. 8.7–8.9, 9.9, 9.10) to ∼40 mm (Figs. 7.5, 7.6, 8.5, 8.6). Different specimens, however, preserve the shortest and longest valve lengths, which range from ∼20 mm (Fig. 9.1–9.4) to 28 mm (Fig. 8.1, 8.2). The ratio of carapace length to total length (from anterior of carapace to posterior margin of telson head) ranges from 0.60 (Fig. 8.5, 8.6) to 1.05 (Fig. 9.9, 9.10), emphasizing the effect of flattening. YPM IP 543312 (Fig. 9.3, 9.4) falls within this range at 0.64, but the outline of the valve is anomalous. The height of the valve is 16 mm, substantially greater than in any other specimen, and the height to length ratio is 0.89, much greater than the mean of ∼0.5. It is likely that this reflects orientation during flattening. The preserved ratio of valve height to length varies from 0.38 (USNM 15314F and YPM IP 543321) (Figs. 7.5, 7.6, 9.9, 9.10) to 0.89 (YPM IP 543312) (Fig. 9.3, 9.4). Values of H/L: USNM 15314F = 0.38; YPM IP 543312 = 0.89; YPM IP 543316 = 0.47; YPM IP 543319 = 0.48; YPM IP 543320 = 0.41; YPM IP 543321 = 0.38; YPM IP 543333 = 0.43; YPM IP 543334 = 0.43; YPM IP 543335 = 0.53; YPM IP 239148 = 0.48 (Mean 0.49, s 0.15, N = 10; omitting the outlier, YPM IP 543312, mean 0.44, s 0.05, N = 9).

Preservation.—We use the terminology introduced by Whittington (1971) for Burgess Shale specimens to describe variation in the orientation of specimens of Vermontcaris montcalmi n. gen. n. sp. to bedding. Whittington (1971) distinguished lateral, parallel (dorso-ventrally flattened), and oblique (intermediate) specimens. Vermontcaris montcalmi n. gen. n. sp. is flattened laterally or slightly obliquely (Figs. 7, 8.1–8.6, 8.8–8.10, 9), indicating greater hydrodynamic stability of carcasses in this attitude, presumably because the body was relatively narrow and the valves were not highly convex. YPM IP 239153 (Fig. 8.7), which appears to be flattened dorso-ventrally, as indicated by the narrow carapace and lack of evidence of appendages projecting laterally beyond the trunk, is an exception (a poorly preserved specimen on USNM 15314 G is similar). Specimens show little direct evidence of decay and disarticulation. The carapace is flipped forward in some examples (Figs. 8.5, 8.6, 8.8–8.10, 9.5–9.10) in an attitude that may correspond to the onset of molting (specimens of Canadaspis perfecta [Walcott, 1912] [Walcott, 1912a] from the Burgess Shale show a similar configuration; Briggs, 1978a, p. 475). All the specimens in question preserve traces of the gut along the length of the trunk, however, indicating that they represent carcasses; there is no evidence that they are preserved in the act of molting (cf., García-Bellido and Collins, 2004; Yang et al., 2019). The gut in YPM IP 543334 is preserved in relief in three sections (Fig. 8.8–8.10); the middle of these has rotated, suggesting that the content was authigenically cemented within the carcass. Other specimens preserve the gut in relief (Figs. 7, 8, 9.3–9.10), including examples where it is evident even where it is overlain by the carapace (Figs. 8.1–8.4, 9.3, 9.4), indicating early cementation.

Discussion.—The streamlined carapace, flap-like exopods extending beyond the ventral margin of the carapace, and leaf-like appendages associated with the telson suggest that Vermontcaris montcalmi n. gen. n. sp. was a swimmer, as were some other Cambrian bivalved arthropods (Fu and Zhang, 2011; Legg and Caron, 2014). The flexible trunk (e.g., Fig. 9.5, 9.6) may have aided in locomotion. The multiple segments and resulting large number of exopods may have provided enhanced respiratory capability in response to low-oxygen conditions, such as those in the Franklin Basin (Pari et al., 2021).

Remarks.—The morphology of Vermontcaris montcalmi n. gen. n. sp., which is characterized by an elongate body, bivalved carapace, and multiple short trunk segments, aligns with that of Cambrian stem-group arthropods such as Jugatacaris (Fu and Zhang, 2011) and Pectocaris (Hou, 1999; Hou et al., 2004; Jin et al., 2021). The number of thoracic segments in V. montcalmi n. gen. n. sp. is difficult to determine, but there may be as many as 40. There is a short abdomen of 4 or 5 segments without appendages. The carapace of Jugatacaris agilis from the Chengjiang biota has a dorsal keel or blade-like structure, in contrast to that of V. montcalmi n. gen. n. sp. The trunk comprises 55–65 segments, including 3–5 apodous segments at the posterior end (Fu and Zhang, 2011). Fu and Zhang (2011, p. 574) did not designate these posteriormost segments as a separate abdomen, acknowledging the difficulty of drawing a distinction between it and the rest of the trunk. The trunks of Pectocaris spatiosa, P. eurypetala (Hou and Sun, 1988), and P. inopinata Jin et al., 2021, have ∼50 segments, but vary in the number that lack appendages: the first two have ∼4, whereas the last, in contrast to V. montcalmi n. gen. n. sp., has 11 or 12 making up a distinct abdomen. The telson of P. spatiosa Hou, 1999, is unknown (Hou, 1999). The telson of P. inopinata bears a pair of lateral appendages distally, whereas P. eurypetala has a median as well as lateral flukes (Hou et al., 2004), similar to the arrangement in Odaraia alata Walcott, 1912 (Walcott, 1912a). The lateral appendages in V. montcalmi n. gen. n. sp., in contrast, appear to articulate with the pretelson segment. Phylogenetic analyses (Legg et al., 2012; Legg and Caron, 2014) resolved Pectocaris and Jugatacaris within a paraphyletic grade of bivalved forms, which are part of ‘Upper’ stem-Euarthropoda at the base of the arthropod stem (Ortega-Hernández, 2016). Bivalved forms with multisegmented elongate bodies likely represent a symplesiomorphic morphology that consistently falls within this clade (J. Yang et al., 2016). Aria and Caron (2017) omitted Jugatacaris from their analysis but regarded it as a putative member of the hymenocarines, which resolved as stem mandibulates (Vannier et al., 2018, obtained hymenocarines as crown mandibulates). These several alternative placements might apply equally to V. montcalmi n. gen. n. sp., but the lack of preserved details of the appendages prevent its position from being resolved further.

Type species.—Herpetogaster collinsi Caron, Conway Morris, and Shu, 2010, Burgess Shale and Stephen Shale formations, Yoho and Kootenay National Parks, British Columbia, Canada.

2021 Herpetogaster?; Pari et al., p. 693, fig. 2H.

Description.—YPM IP 239054 (Fig. 10.1, 10.2) is curved sharply through an angle of ∼30°. The shorter, wider portion terminates in an array of tentacle-like structures, which are too poorly preserved to reveal details, and presumably represents the head. The bend coincides with a wide area of gut fill that extends to the outer convex margin and may represent the stomach. There is no unequivocal trace of the gut along the length of the trunk in either direction beyond this fill, and therefore no evidence of the position of the mouth or anus. The longer portion of the trunk ends in a blunt round, presumably posterior, termination. Wrinkles parallel to the margin near this posterior end indicate compaction of relief. The margins of the trunk are not clearly defined everywhere, and there is no evidence of segmentation. The specimen is ∼80 mm long.

Material.—YPM IP 239054.

Discussion.—Although poorly preserved, the Parker Formation specimen preserves characters that align with Herpetogaster, including the worm-like body, prominent gut trace (possibly the stomach), and array of possible tentacles (Pari et al., 2021). The compaction wrinkles are consistent with a tough, flexible integument (Caron et al., 2010, p. 4). The Parker Formation specimen differs from the lower Cambrian genus Phlogites from the Chengjiang Lagerstätte of China, to which Herpetogaster has been compared (Caron et al., 2010), in lacking a long stolon continuous with the trunk, although such a structure might be concealed by matrix. It is somewhat larger than previously reported specimens of Herpetogaster from other localities.

Caron et al. (2010) and Kimmig et al. (2019) provided comprehensive discussions of the possible deuterostome affinity of Herpetogaster. Caron et al. (2010) assigned it to an unranked stem group, which they named cambroernids (without a formal cladistic analysis) lying somewhere on the branch leading to Echinodermata and/or Hemichordata; Kimmig et al. (2019) followed their lead. The Parker Formation specimen potentially extends the geographic range of the genus to the opposite margin of the paleocontinent Laurentia.

Remarks.—We tentatively assign the single specimen, YPM IP 239054, to Herpetogaster, a genus that was first described by Caron et al. (2010) on the basis of remarkably preserved material from the Burgess Shale and Stephen Shale formations (Cambrian Miaolingian, Wuliuan). The type species, H. collinsi Caron, Conway Morris, and Shu, 2010, is a soft-bodied animal with a tentaculate feeding apparatus and worm-like body, which was attached to the substrate by a stalk or stolon. Herpetogaster collinsi also has been reported from the Ruin Wash Lagerstätte (Cambrian Series 2, Stage 4) in the Pioche Formation of Nevada (Kimmig et al., 2019), which is similar in age to the Parker Formation. The apparent arrangement of tentacles in the Parker Formation specimen is very different from that in H. collinsi, but more similar to that in a second species, H. haiyanensis Yang et al., 2020, from the Chengjiang biota (Yang et al., 2020). Confident assignment of the Parker Formation taxon will require discovery of additional specimens.

Type species.—Emmonsaspis cambrensis (Walcott, 1890), Parker Formation, Parker Quarry, Franklin County, Vermont, USA.

1886 Diplograptus ? simplex Walcott, part, p. 92, pl. 11, fig. 4a (non fig. 4).

1889 Phyllograptus? simplex Walcott, part, p. 388.

1890 Phyllograptus? cambrensis (Walcott), part, p. 604, pl. 59, fig. 3 (non fig. 3a).

1938 Emmonsaspis cambriensis (Walcott); Resser and Howell, part, p. 233, pl. 9, fig. 1 (non figs. 2–4).

1993 Emmonsaspis cambrensis (Walcott, 1890); Conway Morris, p. 599, pl. 1, figs. 1, 2.

2014 Metaspriggina sp.; Conway Morris and Caron, Extended data fig. 6a–d.

2021 Metaspriggina; Pari et al., 2021, p. 693, fig. 2E.

Holotype.—USNM 15314A, Parker Formation, Parker Quarry, Franklin County, Vermont, USA. Cambrian Series 2, Stage 4 (Resser and Howell, 1938, pl. 9, fig,1).

Description.—The original specimens are incomplete, but USNM 15314A, which is the largest, preserves ∼50 myomeres (Conway Morris and Caron, 2014, Extended data fig. 6b). The eye is evident in USNM 15314C. USNM 15314A preserves dark ventral features (interpreted by Conway Morris and Caron, 2014, Extended data fig. 6b, as the heart and liver). YPM IP 239155 preserves evidence of the eyes (Fig. 10.3, 10.4) and even a distinct central area (interpreted as the eye lens by Conway Morris and Caron, 2014, fig. 1). YPM IP 239155 preserves evidence of ∼50 myomeres (Fig. 10.3), but the distal extremity is not evident. A dark area (Fig. 10.3) is preserved in a similar position to that interpreted as the liver by Conway Morris and Caron (2014). YPM IP 543315 (Fig. 10.5) is unusual in being preserved as a rust-colored coating, iron oxide presumably after pyrite, that obscures finer details. USNM 15314A was at least 40 mm long and 10 mm high. YPM IP 543315 (Fig. 10.5) was at least 45 mm long. YPM IP 239156 and YPM IP 239155 (Fig. 10.3, 10.4) were at least 35 and 40 mm long, respectively.

Material.—USNM 15314A, holotype, USNM 15314B, C, YPM IP 239154-239156, YPM IP 543315, YPM IP 543323.

Discussion.—Conway Morris and Caron (2014) referred the original Parker Formation specimens (well illustrated in their Extended data fig. 6) to Metaspriggina spp. without comment. Metaspriggina was not named until 1993, when Simonetta and Insom erected a new genus and species from the Burgess Shale, M. walcotti Simonetta and Insom, 1993, which was subsequently shown to be a chordate (Conway Morris, 2008). We retain Emmonsaspis as the generic name for the Parker Formation taxon, noting that it has priority. Conway Morris and Caron (2014, p. 419) observed that the Parker Formation specimens differ in being less slender and having myomeres with a more angular closure; they also lack evidence of gill bars and zig-zag shaped myomeres. These differences, unless taphonomic, might justify assignment to a separate genus. Otherwise Metaspriggina is a junior synonym of Emmonsaspis.

Remarks.—Resser and Howell (1938) erected Emmonsaspis to accommodate three specimens of this taxon and six specimens that we assign herein to the new taxon, Vermontcaris montcalmi n. gen. n. sp., all from the Parker Quarry Lagerstätte, noting (1938, p. 233) that the morphology “of this animal argue[s] strongly for its reference to the chordates.” They designated all specimens under USNM 15314 as cotypes. Conway Morris (1993) initially considered Emmonsaspis cambrensis to be a frond-like form similar to some Ediacaran animals. Conway Morris (1993, p. 599) provided a full synonymy of the taxon observing that only the first specimen figured by Resser and Howell (1938, pl. 9, fig. 1), USNM 15314A, is attributable to E. cambrensis (Walcott, 1886, pl. 11, fig. 4a; 1890, pl. 59. fig. 3; Conway Morris, 1993, pl. 1, fig. 1). He also noted that the other specimens represent a bivalved arthropod (here described as V. montcalmi n. gen. n. sp.). He therefore designated USNM 15314A as holotype (Conway Morris 1993, p. 600) (i.e., lectotype; ICZN Article 74.1), noting its association with two other less well-preserved specimens (USNM 15314B, C). Conway Morris and Caron (2014, Extended Data fig. 6) reinterpreted E. cambrensis from the Parker Formation as a chordate as part of their redescription of the chordate Metaspriggina walcotti Simonetta and Insom, 1993, from the Burgess Shale (Conway Morris, 2008), based on newly discovered material from the Walcott Quarry and Marble Canyon, assigning the Parker Formation chordate specimens to Metaspriggina sp.