A variety of methodologies exist for preparing plant organs for anatomical studies, many of which are sample dependent (Johansen, 1940; Sass, 1958; Feder and O'Brien, 1968; Carlquist, 1982; Ruzin, 1999; Keating, 2014). Herbaceous materials can be prepared and analyzed using traditional wax-embedding protocols with exceptional results, but hard tissues such as some fruits and seeds can produce less-than-optimal results when prepared with the same methods, and frequently fail to produce publishable data. Traditionally, hydrofluoric acid (HF) has been employed as a softening agent and to remove silica bodies in various organs (e.g., Liao and Wu, 2000), but HF is incredibly caustic, may be difficult to access due to health and safety concerns, and can take more than a month to prepare materials (see Carlquist, 1982 for details). Boiling or autoclaving have also been proposed as effective softening techniques, but highly sclerifled tissues common in fruits and seeds remain notoriously difficult to section (Ruzin, 1999). Carlquist (1982) detailed a protocol originally described by Kukachka (1977) using the softening agent ethylenediamine (ETD) with great success to soften wood prior to embedding, but noted that this technique has limitations when working on materials with a mixture of cell types. In the author's own experience preparing fruits and seeds for anatomical studies, many hardened and mature tissues fail to produce the desired results using the ETD method, which warranted the development of an alternative preparation technique.
In preparing fossil plant material for anatomical studies, it is common practice to embed anatomically preserved fossils in a clear resin and then grind and polish the specimens for viewing with reflected light microscopy (Jones and Rowe, 1999). Alternatively, fossils may be prepared by producing serial thin sections or “wafers” of material using a diamond wafering blade on a lapidary saw (Stein et al., 1982; Hass and Rowe, 1999). These wafers are then mounted onto a microscope slide, ground and polished, and observed using reflected or transmitted light when the specimen is ground thin enough (<1 mm; Benedict et al., 2008). Results of this technique show fine anatomical details of a variety of fossil organs from various ages and sediments (e.g., Manchester, 1994; Hass and Rowe, 1999; Pigg et al., 2004, 2008, 2014; Rothwell and Ash, 2015).
At present, it appears that this wafering technique has not been applied and described in detail for use with nonfossil plant materials. It is the aim of this paper to describe, in detail, the use of the wafering technique modified for extant fruit and seed material.
METHODS AND RESULTS
Producing slides of fruits and seeds for anatomical studies can be difficult due to the often heterogeneous nature of the various tissues in these plant organs. The method proposed here is a modification of techniques used in studying fossil plant material, including embedding techniques (Stein et al., 1982; Jones and Rowe, 1999), and thin sectioning and wafering techniques (Stein et al., 1982; Hass and Rowe, 1999).
The process involves three fundamental steps: (1) embedding specimens in Liquid Bio-Plastic synthetic resin (Ward's Science, Rochester, New York, USA), (2) creating thin sections on a low-speed lapidary saw, and (3) grinding and polishing specimens to obtain anatomical details (Appendix 1). Liquid Bio-Plastic synthetic resin has been used traditionally to prepare whole mounts of specimens for study and teaching (Burger and Seif, 1979), but it is also an ideal resin to create thin sections as it is completely transparent and hard when cured. The resin is mixed with its catalyst (see Table 1 for ratios) and poured as three separate layers, with the specimen embedded in the middle layer (Fig. 1A–F). First, a thin supporting layer (2–3 mm thick) is created to provide a base for the specimen to rest on (Fig. 1A), then the specimen-retaining layer is created to secure the specimen in the desired orientation (Fig. 1B), and a covering layer (2–3 mm thick) is poured to seal the specimen in the block (Fig. 1C).
Liquid Bio-Plastic reference chart for embedding specimens. The proportions are adapted from Ward's Science (n.d.).
The resin blocks with the specimens adequately embedded are then cut into serial thin sections, or wafers, using a lapidary low-speed saw with a diamond cutting edge (e.g., IsoMet Low Speed Saw; Buehler, Lake Bluff, Illinois, USA), which uniformly cuts through the specimen and resin to yield wafers 0.5–1.0 mm thick, depending on the type of blade and lapidary saw used. The wafers are then cleaned, dried, and affixed to a traditional microscope slide using any standard mountant typically used to affix fossil wafers to slides (e.g., Manchester, 1994 [Elmer's epoxy; Elmer's Products Inc., Westerville, Ohio, USA]; Pigg et al., 2004 [UV-cured adhesive, UV-154; T.H.E. Company, Lakewood, Colorado, USA]; Rothwell and Ash, 2015 [Devcon 5 Minute Epoxy; ITW Devcon, Danvers, Massachusetts, USA]). The specimens are then photographed to document general morphology and three-dimensional features that could be lost when grinding specimens for transmitted microscopy. The specimens are then taken through a series of coarse- to fine-grit sandpaper or polishing papers (e.g., 3M Wetordry 6-pc assorted 1–30-µm aluminum oxide/silicon carbide polishing papers; 3M, St. Paul, Minnesota, USA) and ground down to a minimal thickness (<100 µm). Upon completing grinding and polishing, specimens can be made into permanent slides by adhering a coverslip with a long-term mounting medium (e.g., Eukitt; Sigma-Aldrich, St. Louis, Missouri, USA; see Ruzin  for a list of common mounting media or Brown  for a review of various media) and then photographed using transmitted light microscopy (Fig. 2A–H). The author has prepared a variety of plant specimens in this manner for approximately 10 yr and no deformation of tissues as a result of dehydration has been observed; these slides should last as long as traditionally prepared microscope slides.
The major advantages of this technique include: (1) a decrease in processing time (traditional ethylenediamine embedding can take 1–2 wk to embed material, whereas this protocol takes a maximum of 3 d); (2) it does not involve the use of caustic softening agents; (3) it does not remove any chemicals or structures from the specimens (e.g., silica bodies, tannins, lipids) that are commonly removed with polar or nonpolar solvents used in traditional wax-embedding; and (4) it circumvents the breaking or tearing of hard or brittle tissues by avoiding the use of a blade to slice through a tissue sample, which is a common problem in almost any other traditional histological technique. It has been successfully used on fruits and seeds of gymnosperms and angiosperms, including Ephedraceae (Ickert-Bond and Rydin, 2011), Fabaceae (Taylor et al., 2009), Hamamelidaceae (Benedict et al., 2008; Fig. 2G–H; Table 2), Juglandaceae (Taylor et al., 2009), Meliaceae (Pigg et al., 2014), Sapindaceae (Fig. 2D–F; Table 2), and Zingiberales (Benedict, 2012; Benedict et al., 2015; Fig. 2A–C; Table 2), and could potentially be used on any organ that is exceptionally brittle or hard.
List of specimens sampled and voucher information.
Many methods exist for producing thin sections of various plant tissues for anatomical studies, but few methods exist for highly sclerified and heterogeneous tissues of fruits and seeds. The technique outlined here effectively prepares fruit and seed tissues from a variety of gymnosperm and angiosperm families for which traditional embedding/sectioning methods have failed. It has the advantage of dramatically decreasing the processing time of materials from weeks to months, to a maximum of three days. It eliminates the need for polar and nonpolar solvents, which leaves the chemical composition of materials processed intact. It also avoids tearing hard or brittle tissues by removing the need for a sharp blade to section material, which can damage the sectioning apparatus or the desired tissues.
 The author would like to thank K. B. Pigg, M. W. Taylor, S. M. Ickert-Bond, and F. Farrugia for advice and material to prepare, and K. B. Pigg and S. Y. Smith for reviewing this work and offering suggestions. The author would also like to thank the editor and three anonymous reviewers for their constructive and helpful comments. This paper is dedicated to the late M. Tcherepova, who first introduced the author to wafering techniques of fossil plants.