Leaf and inflorescence evidence for near-basal Araceae and an unexpected diversity of other monocots from the late Early Cretaceous of Spain

Abstract

Phylogenetic analyses imply that monocots were a key group in the early radiation of angiosperms, yet they are much rarer than other major clades in the Early Cretaceous macrofossil record. Here we describe a well-preserved leaf and several inflorescences related to the near-basal monocot family Araceae and abundant monocot leaves of uncertain affinities from two latest Albian localities in northeastern Spain. Orontiophyllum ferreri sp. nov. has a multistranded midrib, several orders of parallel-pinnate veins, two orders of transverse veins, and paracytic-oblique stomata. This suite of characters (but with both anomocytic and paracytic-oblique stomata) is characteristic today of Orontium in the near-basal araceous subfamily Orontioideae, and later Cretaceous and early Cenozoic leaves assigned to Orontiophyllum have similar architecture. Sedimentology and anatomy suggest a (semi)aquatic ecology. Other monocot leaves at the same locality are linear and parallel-veined but have similar stomata. Although anomocytic stomata have been proposed as ancestral in monocots, O. ferreri, the associated linear leaves, Albian-Cenomanian cuticles from Australia and Portugal, and extant data are consistent with the hypothesis that variable paracytic-oblique stomata are ancestral. Turolospadix bogneri gen. et sp. nov., from the other locality, includes spadices of ebracteate flowers with four tepals, a central gynoecium, and a long stipe (vs a spathe attached just below the fertile zone as in most Araceae). Phylogenetic analyses indicate that the character combinations seen in O. ferreri and T. bogneri are ancestral for Araceae, and they could be either sister to Araceae or nested within a basal grade of the family. Together with fossils from the Aptian-Albian of Brazil and Portugal, the Spanish fossils indicate that Araceae are among the oldest extant monocot families, but they were associated with diverse linear-leaved monocots of uncertain affinities.

Lemnaceae and Orontiaceae Are Phylogenetically and Morphologically Distinct from Araceae

“…A morphological phylogenetic analysis of extant Arales [5] identified several features that are diagnostic or nearly so for Orontiaceae, including a collenchyma type [112] that is found only in Lysichiton and Symplocarpus, coincidentally the same two Orontiaceae genera whose inflorescence spathes otherwise make them appear superficially more similar to Araceae s.s. (petiole collenchyma is absent in Gymnocarpus and Orontium). Perhaps most noteworthy, there are leaf shape and venation patterns that can enable confident identification of Orontiaceae species, even in fossil material [113,114]. Thus, under more careful examination, there are more than a few characters that contradict the apparent similarity between Araceae s.s. and Orontiaceae.…”

Fossil evidence for Paleocene diversification of Araceae: Bognerospadix gen. nov. and Orontiophyllum grandifolium comb. nov.

“…The matrix was developed from a slightly modified, unpublished version of the morphological matrix published by Cusimano et al (2011; available for download at http:// cate-araceae.myspecies.info/under "Data Sets"), developed as part of the study of Sender et al (2019) and with updates provided by Dr. James A. Doyle, University of California, Davis. That matrix was further modified by (1) rearranging the order of the characters; (2) rewording some of the characters and character states to improve clarity; (3) recoding character 68 for Spathiphyllum Schott from 0 to 0,1; and (4) either removing or adding extinct taxa in several combinations.…”

“…These are (Analysis 1) the slightly modified matrix supplied by Doyle with the extinct species Turolospadix bogneri Sender, J.A. Doyle, Upchurch, Villanueva-Amadoz & J.B. Diez (2019) removed to establish the general topology of the tree as analyzed using only extant taxa, and to test the level of concordance to the relationships among living species inferred by previous studies (i.e., Mayo et al, 1997Mayo et al, , 2013Cabrera et al, 2008;Cusimano et al, 2011;Sender et al, 2019) and summarized by Les (2020) before the fossil taxa were added; (Analysis 2) the same matrix as Analysis 1 with Turolospadix bogneri included; (Analysis 3) the same matrix as Analysis 2 with the aroid leaf morphotaxon Orontiophyllum austriacum (J. Kvaček & A. Herman) J. Kvaček & S.Y.…”

“…Orontiophyllum ferreri Sender, J.A. Doyle, Upchurch, Villaneuva-Amadoz & J.B. Diez, from the Early Cretaceous (Albian) of Teruel Province, Spain, has "three or four" orders of primary veins and two orders of transverse veins (Sender et al, 2019) (Appendix S1, Table S1). Brown (1962) referred three fragmentary monocot leaves from the Paleocene of North Dakota to Alismaphyllites grandifolius (USGS 312893, three specimens; Appendix S3, Fig.…”

The Early Cretaceous Mesofossil Flora of Torres Vedras (Ne of Forte Da Forca), Portugal: A Palaeofloristic Analysis of an Early Angiosperm Community

“…of the Spathiphylleae. To our knowledge this assignment has not been challenged and the presence of fossil aroids in the Early Cretaceous is now well established (Friis et al 2004, Sender et al 2018. The pollen wall of Mayoa is unlike that of superficially similar ephedroid grains.…”