On a new crested pterodactyloid from the Early Cretaceous of the Iberian Peninsula and the radiation of the clade Anhangueria

The pterosaur record from the Iberian Peninsula is mostly scarce and undefined, but in the last few years some new taxa have been described from different Lower Cretaceous sites of Spain. Here we describe a new genus and species of toothed pterodactyloid pterosaur from the Barremian of the Iberian Peninsula, Iberodactylus andreui gen. et sp. nov., that shows a close and rather unexpected relationship with Hamipterus tianshanensis from China. A review of the phylogenetic relationships of the Anhangueria reveals a new family of pterodactyloid pterosaurs, the Hamipteridae fam. nov. being recovered as sister-group of the Anhangueridae. This latter clade can be in turn divided into the new clades Anhanguerinae and Coloborhynchinae. The close relationships of Iberodactylus and Hamipterus shows an interesting palaeobiogeographical correlation between the Chinese and Iberian pterosaur faunas during the Barremian (Lower Cretaceous). The discovery of Iberodactylus strongly suggests that the clade Anhangueria has clear ancestral ties in eastern Laurasia.


Extended description
The specimen MPZ-2014/1 consists of a partial rostrum of approximately 198 mm in length, composed of the anterior part of the premaxilla with a partially preserved premaxillary crest, and a fragment of the maxillary bone including several fragmentary teeth (Fig. 2). The specimen preserves its original three-dimensional shape, although it is somewhat crushed, partly eroded, and several broken bone fragments are lost. The posterior transverse section of the specimen is somewhat triangular in outline, whereas the anterior region exhibits a compressed contour. The frequent fractures of the specimen and the eroded bone surfaces reveal an external thin layer of cortical bone of 1.5 mm, enclosing a delicate trabecular tissue.
Premaxilla. The premaxilla is represented by an anterior fragment of the bone comprising the premaxillary tip up to the posterior broken border, a little behind the level of the eight alveoli. The anterior dorsal margin of the premaxilla is not straight, but slightly curved.
There is no clear suture between the premaxillae and the maxillae, even though a faint groove could be recognized, which has been interpreted as a suture in other anhanguerian pterodactyloids 26 50 . In this pterosaur it was suggested that the suture-like structure, visible between the premaxillary crest and the premaxillae-maxillae, may have separated the soft tissue part from the rest of the skull 49 . This structure seems to be similar in shape and anatomical position to the faint groove observed in Iberodactylus andreui, which presents the same depth of the premaxillary crest sulci. Such sulci could be interpreted as a trait related to the attachment of the rhamphotheca, as in the case of some extant birds 51 .
Regarding the tiny holes located at the anterior margin of the premaxillary crest, their nature is unclear, but some considerations can be made. Firstly, there is no evidence to indicate these structures are originally part of the specimen. They are asymmetrical in number, shape, and position. Furthermore, they do not seem to be of pathological nature as the μCT scan revealed no bone remodelling, resorption, or changes in density (Fig. S1). Therefore, their origin is more likely to be taphonomic in nature. Though hard to ascertain, we suspect that action of osteophagic organisms could be a likely explanation [52][53][54][55] . Nevertheless, the preservation of cortical bone at the surface of the holes -even though this tissue is not present internally-call this interpretation into question. Maxilla. The maxilla is only partially known by a fragment comprising its anteriormost part, extending backwards behind the level of the 7th and 8th alveoli. Albeit with some distortion, the bone preserves most of its three-dimensional shape. It is noticeably crushed at the level of the anterior alveoli, and some of them have been badly eroded along the lateral sides of the bone. Part of the palatal plate is preserved in the specimen. Alveolar row and teeth. Up to fifteen complete or fragmentary dental alveoli are visible on the ventrolateral edge of the rostrum. Alveolus one to eight are preserved on the right side of the specimen, whereas the alveoli one to five, and the seventh and eighth ones, can be recognized on the left side. The sixth alveolus of the left side has been completely eroded away. The first two alveoli are placed on the anterior margin of the rostrum and seem to be forwardly directed. The remainder alveoli face laterally.
Spacing between alveoli is slightly concave and the alveolar lateral walls are clearly convex. They are small and elliptical in shape, between 5 and 9 mm in diameter labiolingually and between 9 and 20 mm anteroposteriorly (See Table S1).
Teeth are set in individual sockets, and not all of them are completely exposed.
Dentition seems heterodont in size and shape, as there is a noticeable variation of basal crown measurements along the tooth row (Table S1). The bases of the crowns are elliptical in shape with the long axis parallel to the sagittal plane of the bone. All the teeth preserved in situ have their crowns broken. They correspond to the right fourth, fifth and seventh teeth, with the eighth tooth on the left side.   0 -femur about twice or longer than metacarpal IV (fe/mcIV > 2.00) 1 -femur longer but less than twice the length of metacarpal IV (1.00 < fe/mcIV < 2.00) 2 -femur about the same length or shorter than metacarpal IV (fe/mcIV < 1.00) 141. Metatarsal III, proportional length relative to tibia length (Kellner 56 (Table   S3). As the size of the larger scapula is proportional to the partial wing they most likely belongs together, and not the smaller scapula. The same is true for the large skull, based on the proportions between the skull and wing elements in Anhanguera piscator and AMNH 22555. We thus tentatively associate here the large skull (IVPP 18931.3) to the partial wing.
Estimates for the total wingspan of IVPP 18931.3 are given in Table S4 based on correlations to the wingspan and measures of each element in Anhanguera piscator, AMNH 22555 and AMNH 22552. In this way, based on Tables S2-4, we can conclude that a Hamipterus tianshanensis individual with a distance between the first and sixth pairs of alveoli of 96.9 cm probably had a wingspan between 3.05 and 3.29 meters.
Scaling this wingspan range to the distance of 119 cm between the first and sixth pairs of alveoli seen in Iberodactylus andreui results in 3.75 and 4.04.  . 22(1), 196-199 (2002).