Carya leroyii, Wheeler & Manchester & Baas, 2023

Wheeler, Elisabeth A., Manchester, Steven R. & Baas, Pieter, 2023, A late Eocene wood assemblage from the Crooked River Basin, Oregon, USA, PaleoBios 40 (14), pp. 1-55 : 20-22

CARYA LEROYII SP. NOV.

FIG. 10A–I

Diagnosis— Growth rings present. Wood semi-ring-porous. Vessels solitary and in radial multiples of 2–3. Perforations simple, intervessel pits crowded alternate medium to large; vessel-ray parenchyma pits similar in size and shape to intervessel pits. Non-septate fibers. Axial parenchyma scanty paratracheal and in narrow bands throughout the growth rings. Rays 1–3-seriate, heterocellular with procumbent body cells and 1–3 marginal rows of square-upright cells. Crystals in chambered axial parenchyma strands, some crystal containing cells enlarged.

Holotype — UF 278-84908 .

Occurence— Dietz Hill ( UF 278).

Etymology— Named in honor of J. F. Leroy in recognition of the importance of his studies of extant and fossil Juglandaceae .

Description —Growth rings present, marked by radially flattened fibers and differences in latewood and ear- lywood vessel diameters of successive growth rings ( Fig. 10A–C). Semi-ring-porous, a slight tendency to diagonal arrangement in some regions ( Fig. 10A).

Vessels solitary and in radial multiples of 2–3, average tangential diameter of earlywood vessels 229 (44) µm, range 136–294 µm; perforations exclusively simple ( Fig. 10H), intervessel pits alternate ( Fig. 10D), ca. 10–12 µm; vessel-parenchyma pits similar in shape and size to in- tervessel pits ( Fig. 10E); vessel element length averages for 451 ( SD =70) µm, range 339–565 µm; thin-walled tyloses present ( Fig. 10B, C, F, G, I).

Fibers non-septate, thin-walled, pits not observed.

Axial parenchyma scanty paratracheal, in tangential bands one to two cells wide throughout the growth ring, bands more closely spaced in the latewood than in the earlywood ( Fig. 10A, B); strands without crystals of 4–8 cells ( Fig. 10G, H).

Rays 1–3-seriate ( Fig. 10 G–H). Multiseriate rays heterocellular, with procumbent body cells and mostly with 1–3 marginal rows of square to upright cells, occasionally more ( Fig. 10E, F); uniseriate rays composed of mostly upright cells, total multiseriate ray height averages 438 ( SD =92) µm, range 283–544 µm; 4–6 per mm.

Crystals common, in chambered axial parenchyma strands, sometimes in much enlarged axial parenchyma cells within an axial parenchyma strand ( Fig. 10G–I).

Comparisons with extant woods— Affinities with Juglandaceae are indicated by the combination of semi-ring-porosity, vessels solitary and in short radial multiples, narrow continuous lines of axial parenchyma, medium-sized alternate intervessel pits, vessel-ray pa- renchyma pits similar in size to intervessel pits.

Its combination of features, including absence of scalariform perforations (a characteristic of Engelhardieae Mann., 1978), is consistent with placement in the Juglandoideae Eaton (1836) (Wheeler, et al. 2022). Our sample is missing the pith, so we cannot confirm that it was solid, as expected in Carya Nuttall (1818) , or septate which characterizes Juglans L. (1753), Pterocarya Kunth (1824) and Cyclocarya Iljinsk. (1953) .

Whether crystals are present and their location are useful features in distinguishing genera and species groups in the Juglandoideae. Crystals are common in this wood so it differs from Pterocarya and the butternut group of Juglans (Asian species of Section Cardiocaryon Dode, 1909 , and the American Juglans cinerea L., 1759, traditionally placed in section Trachycaryon Klotzsch, 1845 ). The tropical black walnuts have crystalliferous axial parenchyma strands, but not in enlarged cells; the north temperate black walnuts have crystals in enlarged axial parenchyma cells, but not in long chains ( Miller 1976). Consequently, Juglans does not seem a good match for UF 278-84908.

The presence of crystals in chambered axial parenchyma and in enlarged axial parenchyma cells suggests affinities with Carya , but not with the North American species of sect. Carya which are ring-porous and have thick-walled fibers ( Stark 1953). Heimsch and Wetmore (1939) noted that crystals in swollen axial parenchyma cells were most common in Carya tonkinensis Lecomte (1921) . Leroy (1950) described the evergreen species Carya sinensis Dode (1912) (synonym: Annamocarya sinensis (Dode) J.- F. Leroy (1950) and compared it to C. tonkinensis . Based on what he acknowledged as limited material, he suggested the two species had similar anatomy. He also noted that the rays in these two species are more heterocellular than in the American species. Müller-Stoll and Mädel (1960) reported another difference for these two species, i.e., they do not have thick-walled vessel elements. However, neither species has crystals in chambered axial parenchyma similar to UF 278-84908.

Neither Heimsch and Wetmore (1939) nor Leroy (1953) gave details about intervessel pitting. The one sample of C. sinensis we had access to has minute-small intervessel pitting (3–5 µm, Wheeler et al 2022); the only other juglandaceous wood with minute intervessel pitting is Rhoiptelea Diels and Hand. -Mazz. (Hand.-Mazz., 1932) ( Withner 1941). Carya cathayensis Sarg. (1916) has crystals in enlarged axial parenchyma cells, but not

Figure 10. Caption on pg. 22.

in chambered axial parenchyma; its heterocellular rays are consistent with earlier observations of Asian species ( Itoh et al. 2022). Unfortunately, we do not have information on the Asian C. hunanensis, C.C. Cheng and R.H. Chang (1979) or C. kweichowensis Kuang and A.M. Lu (1979) . Because of its heterocellular rays and abundant crystals, we suggest that UF 278-84908 is most similar to Asian Carya species.

Comparisons with fossil woods— The two juglandaceous woods of the nearby middle Eocene Nut Beds differ: Clarnoxylon blanchardii Wheeler and Manchester (2002) has prismatic crystals commonly occurring in enlarged ray parenchyma cells; Engelhardioxylon nutbedensis Wheeler and Manchester (2002) has both simple and scalariform perforation plates and prismatic crystals in ray parenchyma. Pterocaryoxylon Müller-Stoll and Mädel (1960) at the nearby Post Hammer locality ( UF 279) lacks crystals entirely ( Wheeler and Manchester 2021). There is at least one species of Carya , possibly two, in the Dietz Hill locality’s carpoflora ( Manchester and McIntosh 2007).

The classic treatment of fossil juglandaceous woods is Müller-Stoll and Mädel’s 1960 paper, Juglandaceen-Hölzer aus dem ungarischen Tertiär des pannonischen Becken, which reviewed the anatomy of the family and all fossil woods assigned to the family by that time. None of the woods they described had the combination of features of UF 278-84908.