A large new leanchoiliid from the Burgess Shale and the influence of inapplicable states on stem arthropod phylogeny

Characterized by atypical frontalmost appendages, leanchoiliids are early arthropods whose phylogenetic placement has been much debated. Morphological interpretations have differed, some of which concern critical characters such as the number of eyes and head appendages, but methodological approaches also have diverged. Here, we describe a new leanchoiliid, Y awunik kootenayi gen. et sp. nov., based on 42 specimens from the newly discovered M arble C anyon locality of the B urgess S hale ( K ootenay N ational P ark, B ritish C olumbia; middle C ambrian). This new morphotype demonstrates the presence of a four‐segmented head in leanchoiliids, along with two small antero‐median eyes in addition to lateral eyes. Y awunik is characterized by a 12‐segmented trunk and a carinate, lanceolate telson adorned with minute spines. The ‘great appendages’ of the animal bear teeth on their two distal rami, which would have enhanced their ability to grasp prey. Attitudes of specimens, resulting from burial at multiple aspects of bedding, suggest the ‘great appendages’ were flexible and capable of antero‐posterior rotation. We also discuss the nature of intersegmental tissues and filaments present within the ‘great appendages’. Our phylogenetic analyses extend the monophyly of leanchoiliids to include H aikoucaris and Y ohoia in a new clade, the C heiromorpha nom. nov. (within H eptopodomera nom. nov.). Other nodes are poorly resolved unless implied weights are used, and in this case, the topology is critically sensitive to the coding prerogative of inapplicable states ( NA s). Both the traditional ‘ A rachnomorpha’ hypothesis ( NA s as additional states) and the more recently favoured ‘ A rtiopoda + C rustacea’ ( NA s as uncertainties) were obtained using the same data set and outgroup. This result stresses, first, the historical importance of polarization over data content in scenarios of early arthropod evolution, and second, a pressing need to investigate the impacts of coding inapplicables, especially given the inflating effect of implied weights.