Independent, structurally distinct transitions to microfruticose growth in the crustose genus Porina (Ostropales, Lecanoromycetes): new isidioid species from south-western Florida

Abstract

Porina is a widely distributed, species-rich genus of crustose, lichen-forming fungi, some with thalline outgrowths that have been recognized as isidia. We studied three taxa with thalli consisting chiefly of ascending isidioid structures occurring on trunks and branches of Taxodium in southwestern Florida, and provide details of their structure with light and electron microscopy. Two of these taxa we describe as new species: P. microcoralloides and P. nanoarbuscula. Genetic sequences (mtSSU) suggest that they are closely related to each other, yet they differ markedly in the size, morphology and anatomical organization of their isidioid branches as well as in the length of their ascospores. In the three Floridian taxa studied, the crustose portion of the thallus is partly endophloeodic and partly superficial, the latter often patchy, evanescent or inconspicuous, and completely lacks the differentiated anatomical organization characteristic of the isidioid structures arising from it. In Porina microcoralloides, the ascendant thallus consists of branched, coralloid inflated structures with phycobiont (Trentepohlia) unicells arranged at the periphery of a loose central medulla. Sparse fungal cells are interspersed and overlie the algal layer in places, but no differentiated cortex is present, leaving phycobiont cells more or less exposed at the surface. In the closely related Porina nanoarbuscula, the isidioid structures are much finer, more densely branched, and composed of a single, central file of roughly spherical Trentepohlia cells surrounded by a jacket of subglobose fungal cells. The ascospores of P. microcoralloides are more than twice the length of those of P. nanoarbuscula. Although thalli of these two Porina species occur in the same habitats and are sometimes found growing alongside each other, phylogenetic analysis of rbcL sequences suggest that they partner with distinct clades of Trentepohlia phycobionts. A third taxon examined, Porina cf. scabrida, is morphologically rather similar to P. microcoralloides, but the ascendant branches are bright yellow-orange, more cylindrical, and corticated by a thin layer of agglutinated fungal hyphae; perithecia were not seen. Analysis of mtSSU sequences places it distant from P. microcoralloides and P. nanoarbuscula phylogenetically. None of the Floridian taxa studied was particularly close to the European isidiate species Porina hibernica and P. pseudohibernica, which appeared as sister to each other in the analysis. While a particular type of isidiose structure may be reliably characteristic of specific taxa, similarities or differences in these structures do not seem to be useful indicators of phylogenetic proximity or distances among taxa. The morphological trends evident in Porina suggest that multiple transitions from crustose to isidioid or microfruticose growth have arisen repeatedly and in quite different ways within this single genus. At least some of the diverse structures treated within the broad concept of isidia may be representative of the developmental pathways by which fruticose growth forms may arise.

Introduction

A majority of lichens develop within the mainly two-dimensional confines of crustose or appressed foliose growth forms closely associated with the substratum surface. Others exploit three-dimensional space by growing and branching upward and outward as fruticose forms. While still only partially explored, there can be significant ecological implications associated with lichen growth forms and overall thallus morphology (e.g. Larson & Kershaw 1976; Pintado et al. 1997; Sojo et al. 1997; Esseen et al. 2015). For example, ascending forms may overgrow and outcompete lower-growing crustose lichens and bryophytes for light (Jahns 1988), as occurs in vascular plant communities. The more extensive surface area of fruticose forms may be more efficient in condensing and absorbing moisture from fog and dew, but will lose moisture more readily when drying conditions prevail (Larson 1981). The energetic costs associated with extensive, supportive mycobiont tissues may also limit the practicality of such thallus forms in warm, humid climates where high respiration rates result in substantial carbon loss at night and under low light conditions (Zotz & Winter 1994). On a broad scale, evolutionary transitions between crustose and fruticose growth forms have occurred many times among lichen-forming fungi, and in both directions. In a few remarkable cases, they may be observed within a single species (e.g. Lecanora swartzii (Ach.) Ach.; Poelt 1989), and apparently correlated with environmental gradients (Weber 1967; Kunkel 1980; Pérez-Ortega et al. 2012). Additionally, most species of the hyperdiverse genus Cladonia have dimorphic thalli, where fruticose axes arising from the horizontal thallus are thought to be homologous with apothecial stipe tissue that later acquired an algal layer and assimilative function (Krabbe 1891; Jahns 1970; Ahti 1982). Nonetheless, a single basic growth form is usually characteristic of a given genus, with relatively few exceptions (e.g. Tehler & Irestedt 2007; Sohrabi et al. 2013). Indeed, lichen growth form may sometimes indicate phylogenetic relationships better than other characters, even at higher taxonomic levels, such as the alectorioid clade within the Parmeliaceae (Crespo et al. 2007). Thus, it appears that where growth form and its strategic implications are concerned, lichen lineages have tended to be conservative, at least at the lower taxonomic levels.

From a principally two-dimensional thallus, however, some degree of vegetative upgrowth and branching in three-dimensional space may also occur. This is evident in the formation of isidia, appendicular organs of diverse structure, development, and phylogenetic origin (Beltman 1978). They are common in foliose and fruticose lichens, occurring more rarely in crustose forms (Jahns 1973), and have long been considered useful as a distinguishing character at species-level (Poelt 1973). Isidia are integral components of the thallus that arise as protuberances of the upper cortex, incorporating fungal and algal tissue as they develop (Hale 1983; Barbosa et al. 2009). The concept of isidia is very broad. Some are transitional with soredia, erumpent symbiotic propagules arising from below the cortex, but isidia are usually distinguished from soredia by their possession of a cortex (Jahns 1973; Beltman 1978). Those isidia that are easily detached may serve, like soredia, as vegetative diaspores (Honegger 1987a; Scheidegger 1995; Zoller et al. 2000), while the basal scars left upon the thallus may aid in CO₂ diffusion, as do pseudocyphellae. Sturdier, less easily detached isidia increase thallus surface area for photosynthesis and condensation, absorption or external storage of moisture (Jahns 1984; Rikkinen 1997; Tretiach et al. 2005); they may also allow more efficient assimilation of CO₂ (Tretiach et al. 2005). Such isidia permit the lichen to take at least partial advantage of three-dimensional space for additional access to light, carbon and/or moisture resources, on a more limited scale than fully fruticose thalli but without relinquishing the extensive contact with substratum resources enjoyed by the underlying thallus. Thus, a lichen's adoption of two-dimensional versus three-dimensional growth patterns has functional significance for its success. Repeated transitions between these growth strategies within a single genus are therefore likely to indicate substantial environmental selection pressures exerted upon morphology.

In south-western Florida, recent examination of lichen communities on Taxodium bark revealed several phenotypically different types of minutely fruticose thallus structures containing Trentepohlia photobionts, often without a well-developed basal crustose thallus. The identities of the lichens, which were initially found without sexual structures, remained mysterious until molecular sequences and eventually perithecia indicated their affinities within the genus Porina. As one of the larger lichen-forming fungal genera, Porina currently includes some 140–300 species, depending on circumscription (McCarthy 2013; Lücking et al. 2017). They occur on bark, rock and leaf substrata, with the highest diversity in humid subtropical and tropical regions. Porina is known as a genus of crustose lichens, with several species described as isidiate (Swinscow 1962; James 1971; Harris 1995; Cáceres et al. 2013; Tretiach 2014; Diederich et al. 2017; Orange et al. 2020; Ertz & Diederich 2022). In the present work, we attempt to better understand the structural and phylogenetic context of such morphological transitions within the genus Porina. We describe and compare the structure of our isidiate/microfruticose Porina collections using light and scanning electron microscopy, and examine molecular markers to determine phylogenetic placement among other Porina species for which sequences are available. Additionally, since unstratified crustose lichens commonly show intracellular haustorial penetrations while fruticose lichens are usually found to have non-intrusive fungal-algal contacts (Tschermak 1941; Honegger 1986), we examine symbiont interfaces with TEM to see how this paradigm might apply in our structurally transitional lichen collections.

Materials and Methods

Sample collection and microscopy

Lichens were collected on Taxodium bark within and near the margins of seasonally flooded groves on the Florida Gulf Coast University campus, at a nearby residential community, at Corkscrew Swamp Sanctuary, and at Corkscrew Regional Ecosystem Watershed (CREW) in Lee County and Collier County, Florida. Voucher/type specimens will be deposited at FLAS (University of Florida), with duplicates at BR and TSB herbaria.

Thalli were examined and photographed with an Olympus SZX12 dissecting microscope equipped with an Infinity 3S camera. Fruticose branches and hand sections of perithecia were wet mounted in tap water, 10% KOH (K), or in Lugol's iodine solution (1% I₂) without (I) or with K pre-treatment (KI), and photographed through an Olympus BX51 compound microscope. Colour reactions of the thallus were studied using K, common household bleach (C), crystals of para-phenylenediamine dissolved in ethanol (PD) and long wave UV (366 nm). Ascospores measurements are indicated as (minimum value–)mean(–maximum value), followed by the number of measurements (n).

Specimens were affixed to SEM stubs with carbon adhesive, coated with gold, and examined with an FEI Inspect scanning electron microscope (Thermo Fisher Scientific, Waltham, Massachusetts).

Specimens selected for embedding were hydrated in Petri dishes with moist filter paper 24 h prior to further processing according to de los Ríos & Ascaso (2002). The samples were then fixed in 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.1) for 3 h at 4–5 °C, with vacuum infiltration for three 10-min periods inside a desiccator during the initial portion of the fixation period. Specimens were then washed three times, 30 min each, with phosphate buffer at room temperature, followed by post-fixation in 1% osmium tetroxide for 5 h in the dark. The post-fixed material was then washed three times in buffer, dehydrated in an ethanol series followed by propylene oxide, then infiltrated in a 1:1 mixture of propylene oxide and Spurr's low-viscosity resin. The following day, the specimens were infiltrated with fresh resin, left for three days in the refrigerator, then polymerized at 60 °C.

Semi-thin sections were cut 1–2 μm thick and stained with toluidine blue. Ultrathin sections 79 nm thick were stained with uranyl acetate followed by lead citrate.

Molecular analyses: DNA extraction, PCR amplification and sequencing

Genomic DNA was extracted from a total of 35 lichen thalli (Table 1), following the CTAB protocol according to Cubero et al. (1999). The Floridian samples used for DNA extraction corresponded to those that were morphologically studied; two samples of Porina hibernica P. James & Swinscow and three of P. pseudohibernica Tretiach, collected in their type localities, were also included. Mycobiont sequences were compared with those available in GenBank. Part of the DNA coding for the small subunit of the mitochondrial ribosome (mtSSU) was amplified using the primers mrSSU1 and mrSSU3R (Zoller et al. 1999). The ITS locus was amplified using the forward primer ITS1F (Gardes & Bruns 1993) and the specific Porina reverse primer ITSPoR (5′ - CCT TGC CTG ATC CGA AGT GAA ACC G - 3′; Orange et al. 2020). Chloroplast DNA corresponding to the large subunit of ribulose-1,5-biphosphate carboxylase (rbcL) was amplified with the primers rbcL803rev and rbcl320 (Nozaki 1995) to check the identity of the photobiont. The PCR conditions for the amplification of the mtSSU locus followed Orange et al. (2020), while those for the rbcL locus followed Muggia et al. (2008, 2010). The PCR products were purified with Mag-Bind® Total Pure NGS; Sanger sequencing was performed by Macrogen Europe, Inc. (Amsterdam) using the forward primers for all loci. The identity of the sequences was checked with a BLAST search (Altschul et al. 1990) in the NCBI database.

Table 1. Porina specimens newly sequenced and included in the phylogenetic analyses of the present study, with their collection and DNA extraction numbers and NCBI Accession codes for the ITS, mtSSU and rbcL marker sequences obtained. NAS and WBS refer to collection numbers of N. A. Sanderson and the first author, respectively; TSB refers to collections accessioned at the University of Trieste Herbarium.

Phylogenetic analyses

The phylogenetic analyses of Porina mycobionts included the newly generated mtSSU and ITS sequences, plus 139 sequences retrieved from NCBI for the mtSSU locus and 38 for the ITS locus. All sequences were aligned in a comprehensive dataset individually constructed for each locus. For those samples for which both ITS and mtSSU sequences were available (i.e. those 15 retrieved from NCBI and nine new samples we collected), a multilocus concatenated dataset was prepared in MEGA11 (Tamura et al. 2021). Coenogonium leprieurii (Mont.) Nyl., C. luteum (Dicks.) Kalb & Lücking, C. pineti (Ach.) Lücking & Lumbsch, Gyalidea praetermissa Foucard & G. Thor and Sagiolechia protuberans (Ach.) A. Massal. were selected as outgroups, according to Orange et al. (2020) and Ertz & Dieterich (2022), for the mtSSU dataset; Porina austroatlantica P.M. McCarthy & Fryday and Porina multipuncta (Coppins & P. James) Ertz, et al. were selected as outgroups for the ITS dataset, according to Orange et al. (2020), and P. austroatlantica was also set as outgroup for the concatenated ITS + mtSSU dataset. The phylogenetic analyses of the photobiont included the newly generated rbcL sequences and 444 sequences retrieved from NCBI, including the genera Trentepohlia and Printzina, while as outgroups the species Batophora oerstedii, Bornetella nitida, Bryopsis hypnoides, Caulerpa prolifera, Halimeda discoidea, H. opuntia, Polyphysa peniculus and Ulva linta were selected (according to Rindi et al. (2009) and Borgato et al. (2022)). The alignments were performed using MAFFT v.7 (Katoh et al. 2002) with MSA algorithm set on 100 bootstrap replicates and G-ins-I as the substitution model, and then manually adjusted in BioEdit v.7.2.5 (Hall 1999).

Maximum likelihood (ML) and Bayesian inference (BI) analyses were run for both the Porina and photobiont datasets on the CIPRES web portal (Miller et al. 2010), using the programs RaxML v. 8.2.12 (Stamatakis 2014) and MrBayes v. 3.2.7a (Huelsenbeck & Ronquist 2001), respectively. The ML analysis used the GTRGAMMA substitution model, with 1000 bootstrap replicates; the BI was carried out setting two parallel runs with six chains over five million generations, starting with a random tree and sampling every 100th step. We discarded the first 25% of the data as burn-in, and the corresponding posterior probabilities (PPs) were calculated from the remaining trees.

The phylogenetic trees were visualized in TreeView v. 1.6.6 (Page 1996). Species level lineages were recognized as those clades that were monophyletic, fully supported, and represented by more than two samples; they were named according to Orange et al. (2020), Ertz & Diederich (2022) and Borgato et al. (2022).

Results

Structural features of the material examined

Porina microcoralloides

The ascending thallus consisted of somewhat irregularly swollen, branching, coralloid microfruticose structures, ranging in colour from yellowish brown to dark brown or dark olive green (Fig. 1A–F & K). The crustose thallus at the base of these structures appeared patchy, often inconspicuous or evanescent (Fig. 1B, C & E). Basal thallus portions comprising a superficial crust consisted of a mixture of scattered fungal hyphae and individual rounded cells or short filaments of Trentepohlia phycobionts, without any discernable organization into discrete tissue layers (Fig. 2A). Material associated with the cell walls of some of the superficial fungal cells formed a chiefly acellular epilayer at the surface of the crustose thallus (Fig. 2A). In other places, the basal thallus consisted of mycobiont hyphae and Trentepohlia cells growing loosely over the substratum (Fig. 3E) and/or occupying empty cell lumina of the bark substratum (Fig. 2B, C & F), from which isidioid structures emerged directly (Figs 2B, 3G & H).

Figure 1. Porina microcoralloides, P. nanoarbuscula and P. cf. scabrida from south-west Florida. Dissecting microscope and whole mounted compound microscope images. A–F, P. microcoralloides. G–I, P. nanoarbuscula. Arrowheads: perithecia. J, section through plant substratum (s) with embedded thallus and underlying perithecium (p) of P. nanoarbuscula. K, P. microcoralloides, isidioid structure whole-mounted in water. L & M, P. nanoarbuscula, isidioid structure whole-mounted in water and aniline blue, respectively. N, P. microcoralloides (lower half of image) and P. nanoarbuscula (upper half of image) growing intermixed. O & P, P. cf. scabrida. Scales: A, B, C & H = 100 μm; D & N = 500 μm; E & G = 200 μm; F, I, O & P = 250 μm; J & K = 25 μm; L & M = 10 μm. (A, WBS 20425.1; B, WBS 20424.6; C, WBS 20424.6; D, WBS 20423.9; E, WBS 20425.4; F, WBS 21501.5; G, WBS 20424.4; H, WBS 20423.2; I, WBS 21421.8; J, WBS 20424.6; L & M, WBS 20423.9a; N, WBS 20423.2; O, WBS 20506.1; P, WBS 21212.7). In colour online.

Figure 2. Sections of resin-embedded thalli of Porina microcoralloides, examined with light microscopy (A–D) and TEM (E–G). A, unstratified crustose thallus on surface of plant substratum (ps) at left; isidioid structure (arrow) with heteromerous anatomy at right, showing algal layer (a) surrounding medulla (m). B, isidioid primordium (arrow) emerging from plant substratum (ps); phycobiont (a) unicells and filaments in primordium and within lumen of dead plant cells below. C, later stage of emergence directly from plant substratum: note stratification of primordium into algal layer (a) and medulla (m). D, section through portion of a mature isidioid structure. E, periphery of isidioid structure, with mycobiont cells (f) interspersed among algal symbionts (a) and partial epilayer of material associated with fungal cell walls (arrowheads). F, lichen symbionts associated within the confines of substratum plant cell walls (pcw). G, intrusive symbiotic contact between mycobiont (f) and phycobiont (a), showing local invagination of algal cell wall and thinning of fungal wall in contact zone. H, perithecium (p) developing with delaminated layers or plant substratum (ps). Scales: A & H = 50 μm; B & D = 20 μm; C, E & F = 10 μm; G = 1 μm.

Figure 3. Scanning electron micrographs of Porina microcoralloides. A–D, views of branching isidioid structures, with algal cells (a) visible at surface. E, crustose mat of loosely organized symbionts (centre) with isidioid structures arising at periphery. F, detail of E showing Trentepohlia phycobionts (a) and associated mycobiont cells (f). G & H, isidioid structures emerging directly from plant substratum: note absence of any crustose thallus upon substratum surface. I, isidioid fragment (i) establishing on substratum; note radiating attachment hyphae (arrows). Scales: A, B & H = 20 μm; C, D & F = 10 μm; E, G & I = 50 μm.

In contrast with substratic thallus portions, the ascending isidioid structures had a distinctly stratified anatomy. They were composed of a central region of sparse fungal hyphae surrounded by a peripheral layer of subspherical algal cells (Fig. 2A, C & D). Some fungal cells were exterior to, as well as interspersed among, the algal symbionts, and a largely acellular epilayer of material associated with these fungal cells was sometimes evident in sections (Fig. 2E), but no organized cortex was present (Fig. 3A–D). Indeed, algal cells were often visible at the exterior surface in scanning electron micrographs (Fig. 3C, D & F). Within zones of intimate symbiont contact, algal cell walls were invaginated to a modest degree around an intrusive protuberance of a fungal cell whose wall often appeared reduced in thickness at the contact point (Fig. 2G).

Perithecia were approximately globose, sometimes somewhat pyriform, and mostly immersed in the plant substratum; they possessed a thick, carbonized wall (Figs 2H, 4A & B). Ascospores were long-bacilliform to needle-shaped, averaging 66 μm × 3.7 μm, with c. 11–13 transverse septa (Fig. 4C–F).

Figure 4. Perithecium and ascospores of Porina microcoralloides (A–F) and P. nanoarbuscula (G–M). A, perithecium. B, melanized perithecial wall tissue. C–E, free ascospores. F, ascospores in ascus (C, live cell, bright field; D–F, in KOH, DIC optics). G, perithecium. H–M, free ascospores (H, live cell, bright field; I–M, in KOH, DIC optics). Scales: A = 50 μm; B & G = 20 μm; C–F = 25 μm; H–M = 10 μm. In colour online.

Porina nanoarbuscula

Crustose thallus portions on the substratum surface were often patchy and of limited extent (Fig. 1G & I). They consisted of subglobose to short filamentous photobiont cells and scattered mycobiont hyphae without stratification or any indication of cortical development (Fig. 5A). A chiefly acellular epilayer of material associated with the cell walls of some superficial mycobiont cells often formed at the upper surface (Fig. 5A). In other places, the surface crustose layer was not developed, and ascending isidioid structures arose from a disorganized mixture of mycobiont and photobiont cells colonizing dead, superficial cells within the plant substratum (Figs 1J, 5B & C). The isidioid structures were exceptionally fine and densely branched, the branches breaking and detaching readily upon mechanical contact (Figs 1H, 6A, B & H). They were each composed of a single central file of more or less globose Trentepohlia cells, surrounded peripherally by relatively swollen, subglobose mycobiont cells (Figs 1L & M, 5B & D, 6C–H). Short chains of these cells were also occasionally seen running along the substratum surface in the vicinity of the isidioid thallus (Fig. 6I). Within differentiated symbiont contact zones, algal cell walls were invaginated to a modest degree around the intrusive protuberance of a fungal cell whose wall often appeared reduced in thickness at the contact point (Fig. 5E).

Figure 5. Sections of resin-embedded thalli of Porina nanoarbuscula, examined with light microscopy (A & B) and TEM (C–E). A, thin unstratified thallus crust on surface of plant substratum (ps). B, isidioid structure emerging from symbionts within substratum; pcw, plant cell wall. C, fungal (f) and algal (a) symbionts among cell walls of plant substratum (pcw). D, portion of isidioid structure showing uniseriate central strand of algal symbiont (a) and surrounding mycobiont cells (f); ih, intrahyphal hypha. E, intrusive symbiotic contact between mycobiont (f) and phycobiont (a), showing local invagination of algal cell wall and thinning of fungal wall in contact zone. Scales: A = 20 μm; B = 10 μm; C & D = 5 μm: E = 1 μm.

Figure 6. Scanning electron micrographs of Porina nanoarbuscula. A & B, densely branching isidioid structures. C, surface layer of subglobose fungal cells. D & E, surface layer, with some deposition of wall associated substances somewhat obscuring the individual fungal cells. F–I, backscattered electron detector images highlighting individual fungal cells of surface layer. H, broken ends of isidioid structures showing central zone (t) normally occupied by a single central file of Trentepohlia cells. I, low-magnification image showing isidioid structures (i) arising from substratum in absence of basal crust; arrows, mycobiont cells overrunning substratum; arrowheads, wall thickenings of plant substratum. Scales: A, B, D, F & I = 20 μm; C, E, G & H = 10 μm.

Perithecia were approximately globose with a thick, carbonized wall, mostly immersed in the plant substratum (Fig. 4G). Ascospores were bacilliform, averaging 29 × 3.3 μm, with 3–5 transverse septa (Fig. 4H–M).

Porina cf. scabrida

The thallus resembled that of P. microcoralloides, but was orangey yellow with somewhat more vertical, cylindrical branches (Figs 1O & P, 7A & B) containing clusters of calcium oxalate crystals. As in P. microcoralloides, phycobiont unicells were arranged in a distinct layer at the periphery of a lax central medulla of mycobiont hyphae (Fig. 7C). Exterior to the algal cells, however, a more developed layer of agglutinated fungal hyphae formed at the surface (Fig. 7D–G); exposed phycobiont cells were not observed at the surface, in contrast with P. microcoralloides. At symbiont contact zones, limited invagination of the algal cell wall was evident, with thinning of both algal and fungal cell walls visible at the point of ‘haustorial’ intrusion (Fig. 7H). Perithecia were not observed in the two collections of this taxon.

Figure 7. Light and electron micrographs of isidioid structures in Porina cf. scabrida. A & B, SEM images showing morphology of branches. C, resin-embedded semi-thin section showing algal layer (a) at periphery of a central medullary cavity (m). D–F, surface layer of agglutinated mycobiont hyphae; no exposed algal cells evident. G & H, TEM images of peripheral portion of structure. G, algal layer (a) and associated mycobionts cells (f), with wall-derived materials (arrows) forming an agglutinating layer among mycobiont cells at the surface. H, symbiont contact showing invagination of algal cell ahead of intruding mycobiont haustorium, and the walls of both symbionts substantially thinned at contact zone. Scales: A & B = 50 μm; D = 25 μm; C, E & F = 10 μm; G = 5 μm; H = 1 μm.

Porina hibernica sample from Great Britain

For comparison, the widely reported isidiate Porina hibernica was also studied. In the collection examined, irregular isidia-like upgrowths emerged from a crustose thallus (Fig. 8A–C). In section, these upgrowths appeared to lack stratification or differentiation of symbionts into discrete layers (Fig. 8C). Mycobiont and phycobionts inhabited dead cells of the plant substratum (Fig. 8C–D), and dead cell walls of the plant substratum were incorporated into the isidia-like upgrowth (Fig. 8C). Slight invagination and substantial thinning of both phycobiont and mycobiont cell walls was evident at sites of symbiont contact (Fig. 8E).

Figure 8. Light and electron micrographs of isidioid structures in Porina hibernica from Great Britain. A & B, SEM images showing outgrowth of irregularly shaped isidia from well-developed crustose thallus. C, semi-thin section of resin-embedded material; emerging isidium with unstratified algal cells (a) and incorporating the cell wall lattice (arrows) of the underlying plant substratum. D & E, TEM images. D, associated mycobiont (f) and phycobiont (a) within empty cells of plant substratum (arrows). E, detail of symbiont contact zone, showing slight invagination of algal cell wall and substantial thinning of fungal cell wall at contact point. Scales: A = 50 μm; B = 25 μm; C = 20 μm; D = 5 μm; E = 1 μm.

Phylogenetic analyses

A total of 29 mtSSU and 13 ITS sequences for Porina and 25 rbcL sequences for their Trentepohlia photobionts were newly obtained in this study. The phylogenetic inferences based on the individual loci ITS and mtSSU of the mycobiont (Figs 9 & 10) were congruent with the recent phylogenetic reconstructions presented by Orange et al. (2020) and Ertz & Diederich (2022) for the genus Porina. The multilocus (ITS + mtSSU) phylogenetic reconstruction (see Supplementary Material Fig. S1, available online) was concordant with the single locus phylogenies.

Figure 9. Phylogenetic hypothesis based on the mtSSU locus; 50% majority-rule consensus tree obtained by Bayesian analysis. ML bootstrap values > 70% shown in bold branches; Bayesian PP values > 0.8 are reported above branches. DNA extraction numbers of the new sequences obtained from P. microcoralloides, P. nanoarbuscula, P. cf. scabrida and an additional south Floridian collection (L3531), as well as P. hibernica and P. pseudohibernica are highlighted in bold. ‘Porina chlorotica’ appears as several distinct clades labeled with letters in parentheses.

Figure 10. Phylogenetic hypothesis based on the ITS locus; 50% majority-rule consensus tree obtained by Bayesian analysis. ML bootstrap values > 70% shown in bold branches; Bayesian PP values > 0.8 are reported above branches. DNA extraction numbers of the new sequences obtained from Porina nanoarbuscula, P. hibernica and P. pseudohibernica are highlighted in bold. ‘Porina chlorotica’ appears as several distinct clades labeled with letters in parentheses.

Among Porina species for which mtSSU sequences were available, P. microcoralloides and P. nanoarbuscula were placed as well-supported sister clades (Fig. 9). The European isidiate taxa P. hibernica and P. pseudohibernica were well-supported clades sister to each other, and not very closely related to P. microcoralloides and P. nanoarbuscula. Rather, they were placed closely to P. collina Orange et al. and P. byssophila (Körb. ex Hepp) Zahlbr. This result is also corroborated by the two-loci analysis (Supplementary Material Fig. S1). The two sequenced samples of Porina cf. scabrida were quite distant from the four aforementioned taxa in an unresolved clade with P. nucula Ach. One last sample of Porina, namely L3135, was placed close to Porina cryptostoma Mont. and two sequences of Porina nucula. This single specimen appeared similar to P. cf. scabrida, and its placement elsewhere was unexpected. We were unable to study it further in the present work.

The ITS phylogeny (Fig. 10) confirmed the sister relationship between P. hibernica (represented by a single sequence) and P. pseudohibernica, and their distance from P. nanoarbuscula, which was sister to P. sorediata Aptroot et al. (represented by a single sequence). However, relatively few ITS sequences were available from GenBank, nor were we able to obtain many from our collections. No ITS data could be obtained from P. microcoralloides, for which several attempts at PCR amplification were unsuccessful.

The Trentepohlia sequences we obtained from the isidiate Porina phycobionts segregate well into separate monophyletic clades. They are all relatively distant from the other Trentepohliaceae, forming a clearly defined major clade in the phylogeny. Phycobiont rbcL sequences from Porina microcoralloides and P. nanoarbuscula thalli indicated that these two mycobionts were each consistently associated with a distinct pool of Trentepohlia strains (Fig. 11; Supplementary Material Fig. S2, available online). Only one sequence was obtained for the Porina cf. scabrida photobiont, and it was placed close to two sequences from free-living Trentepohlia cf. annulata and T. cf. umbrina. A single sequence was also obtained for Trentepohlia sp. from Porina pseudohibernica, which was placed close to one from free-living Trentepohlia annulata and others (OL956825, OL956907) obtained from the lichen Enterographa zonata. The three new sequences obtained from Porina hibernica formed a single, well-supported clade.

Figure 11. Phylogenetic hypothesis based on phycobiont plastidial rbcL locus. 50% majority-rule consensus tree from Bayesian analysis. ML bootstrap values > 70% are shown in bold branches; Bayesian PP values > 0.8 are reported above branches. DNA extraction numbers of the new sequences obtained for Trentepohlia sp. are in bold. Clade numbers in the phylogeny correspond to those assigned in Borgato et al. (2022).

The Trentepohlia sequence obtained from the sample L3135 was placed in Clade 31 sensu Borgato et al. (2022). This clade also included sequences obtained from lichen samples (species of Enterographa, Opegrapha and Porina leptalea) from European temperate forests.

Taxonomy

Porina microcoralloides Ertz, W. B. Sanders, R. Carolis, A. Ríos & Muggia sp. nov.

MycoBank No.: MB 848937

This species resembles Porina coralloidea P. James in its isidiate thallus and its small black perithecia but differs by having (8–)11–13(–15)-septate ascospores of (55–)66.3(–92) × (3–)3.7(–4.5) μm that are less septate (9–11 septa), shorter (40–57 μm) and much broader (9–13 μm) in P. coralloidea.

Type: USA, Florida, Lee County, Fort Myers, Florida Gulf Coast University campus, Cypress swamp north of Parking Garage 3, on Taxodium bark, 20 March 2021, W. B. Sanders 21320.5 (FLAS—holotype).

(Figs 1A–F, K & N, 2, 3, 4A–F)

Thallus consisting of a crustose basal portion from which isidioid structures emerged directly; basal thallus mostly endophloeodal or rarely growing loosely over the bark substratum, ecorticate, consisting of a mixture of individual rounded cells or short filaments of trentepohlioid photobiont cells and scattered mycobiont hyphae without stratification; isidioid structures often abundant, forming dense clusters c. 0.1–1 mm diam. or covering more or less evenly larger areas of the substratum, ascending, richly branched-coralloid, irregularly swollen, 30–45(–50) μm broad and up to c. 350 μm long, easily breaking and detaching upon mechanical contact, yellowish brown to dark brown or dark olive green, each composed of a central region of sparse fungal hyphae surrounded by a peripheral layer consisting of subspherical algal cells c. (5–)7–11(–14) μm diam. interspersed or sometimes covered with some fungal hyphae c. 2–4 μm diam.; without crystals; soralia absent; prothallus inconspicuous, or presence of a black borderline c. 0.2–0.5 mm wide.

Ascomata perithecioid, scattered, rarely two contiguous, subglobose or rarely somewhat pyriform, black, smooth to slightly rugulose, (150–)234(–290) μm diam. (n = 16), c. one third to almost entirely immersed in the substratum, without thallus cover and setae; crystallostratum absent; ostiole apical, usually visible by a tiny black pore or inconspicuous. Proper excipulum dark reddish brown to carbonized all around the hymenium, K± olivaceous black, c. 20–45 μm thick. Involucrellum reduced, appearing as a thickening of the upper part of the excipulum, dark reddish brown or ±purplish brown to carbonized, K± olivaceous black, c. 45–70 μm thick. Periphyses numerous, 10–22 × 1.5–2 μm. Hamathecium hyaline, clear, of thin, simple or sparingly furcate-branched, (1–)1.5–2 μm diam.; paraphyses c. 110–150(–170) μm tall; subhymenium hyaline to pale fawn, 8–15 μm thick. Asci cylindrical-clavate to ±fusiform, c. (75–)100–118 × 13–15 μm (n = 11), 8-spored; ascus apex rounded, without a ring structure. Ascospores hyaline, transversely (8–)11–13(–15)-septate, long-bacilliform to needle-shaped, (55–)66.3(–92) × (3–)3.7(–4.5) μm (n = 25); usually with a gelatinous sheath c. 0.5 μm thick.

Pycnidia not observed.

Chemistry

Thallus and isidia K+ blackish, C−, PD−, UV−. TLC not performed.

Etymology

The epithet refers to the micro-coralloid habit of the thallus consisting mostly of ascending isidioid structures.

Distribution and ecology

The species is known from several localities in south-west Florida (Collier and Lee Co.), where it inhabits the bark of Taxodium within and near the margins of seasonally flooded groves, at low elevation (c. 3–6 m).

Notes

The long, almost needle-shaped ascospores distinguish P. microcoralloides from most isidiate species of Porina, although two non-isidiate taxa with spores at least as long and half the width have been recently described (Aptroot & Cáceres 2014). The British taxon P. hibernica has spores of similar length, but about twice as wide; its isidia, at least in the material we examined, were highly irregular in shape and lacked the stratified anatomy of P. microcoralloides (Swinscow 1962). Phylogenetic distance between these taxa was also evident in their mtSSU sequences (Fig. 9). In Florida, the isidiate taxon Clathroporina isidiifera has a black hypothallus and much shorter, wider ascospores (Harris 1995). A few species of Porina described recently by Ertz & Diederich (2022) from paleotropical Mauritius have isidia that somewhat resemble those of P. microcoralloides; however, they arise from well-developed crustose thalli, have quite different ascospore dimensions, and are placed distantly from P. microcoralloides in the mtSSU sequence analysis (Fig. 9).

Additional specimens examined (all on Taxodium bark)

USA: Florida: Lee County, Fort Myers, Florida Gulf Coast University (FGCU) campus, Cypress dome north of FGCU Blvd near Parking Garage 3, 2020, W. B. Sanders 20423.6 (FLAS, BR), 20423.11b (FLAS); ibid., north of Thalia swamp at centre of dome, 2020, W. B. Sanders 20423.9 (FLAS, BR); ibid., Cypress swamp north of Parking Garage 3, 2021, W. B. Sanders 21501.5 (TSB 44459); ibid., Cypress dome south of campus, near parking garage 2, 2020, W. B. Sanders 20511.2 (FLAS); ibid., Cypress dome between FGCU Blvd and Aquatic Center, 2020, W. B. Sanders 20424.5 (FLAS); ibid., Cypress dome near solar field, FGCU main entrance, 2020, W. B. Sanders 20425.1 (FLAS); ibid., Cypress swamp between swimming pool and FGCU Blvd, 2021, W. B. Sanders 21410.1, 21410.2 (FLAS); ibid., Cypress dome north of main entrance road, 2021, W. B. Sanders 21425.6 (FLAS); ibid., Estero, Grandezza, Cypress grove behind ‘The Studio’ sales centre, 2020, W. B. Sanders 20506.3 (FLAS); ibid., woods between Villa Grande and Grande Estates, 2020, W. B. Sanders 20510.3 (FLAS, BR); Collier County, Naples, Audubon Corkscrew Swamp Sanctuary, 2021, W. B. Sanders 21312.16, 21312.17 (FLAS).

Porina nanoarbuscula Ertz, W. B. Sanders, R. Carolis, A. Ríos & Muggia sp. nov.

MycoBank No.: MB 848938

This species resembles Porina coralloidea P. James by its isidiate thallus and its small black perithecia but differs by having 3–5(–6)-septate ascospores, (20–)28.8(–37) × (3–)3.3(–4) μm; ascospores of P. coralloidea are more septate (9–11 septa) and much larger (40–57 × 9–13 μm).

Type: USA, Florida, Lee County, Fort Myers, Florida Gulf Coast University campus, Cypress swamp north of Parking Garage 3, on Taxodium bark, 3 April 2021, W. B. Sanders 21403.5a (FLAS—holotype).

(Figs 1G–J, L–N, 4G–M, 5 & 6)

Thallus consisting of a crustose basal portion from which isidioid structures emerged directly; basal thallus endophloeodal, ecorticate, consisting of subglobose to short filamentous trentepohlioid photobiont cells and scattered mycobiont hyphae without stratification; isidioid structures often abundant, forming dense clusters c. 0.2–0.6 mm diam. or covering more evenly larger areas of the substratum, ascending, densely branched, cylindrical, fine, 12–16(–20) μm broad and up to c. 200 μm long, easily breaking and detaching upon mechanical contact, orange-brown to dark brown, each composed of single central file of more or less globose Trentepohlia cells c. (6–)8–11 μm diam., surrounded peripherally by relatively swollen, subglobose to slightly elongated mycobiont cells (3–)4–5 μm diam. or 5–6 × 3–4 μm, without crystals; soralia absent; prothallus inconspicuous.

Ascomata perithecioid, scattered, rarely two contiguous, subglobose, black, smooth to slightly rugulose, 155–264.4–350 μm diam. (n = 46), c. two fifths to almost entirely immersed in the substratum, without thallus cover and setae; crystallostratum absent; ostiole apical, visible by a tiny black pore or inconspicuous. Proper excipulum dark reddish brown to carbonized all around the hymenium, K± olivaceous black, c. 18–25 μm thick. Involucrellum reduced, appearing as a thickening of the upper part of the excipulum, sometimes extending slightly laterally when the perithecia is almost entirely immersed in the substratum, dark reddish brown to carbonized, K± olivaceous black, c. 40–50 μm thick. Periphyses numerous, c. 5–30 × 1.5–2 μm. Hamathecium hyaline, clear, of thin, simple, (1–)1.5–2 μm diam.; paraphyses c. 125–130 μm tall; subhymenium hyaline to pale fawn, 14–20 μm thick. Asci cylindrical-clavate to ±fusiform, c. (75–)80–122 × 9–11 μm (n = 12), 8-spored; ascus apex rounded, without a ring structure. Ascospores hyaline, transversely 3–5(–6)-septate, elongate-fusiform to bacilliform, (20–)28.8(–37) × (3–)3.3(–4) μm (n = 54); gelatinous sheath not seen.

Pycnidia not observed.

Chemistry

Thallus and isidia K+ blackish, C−, PD−, UV−. TLC not performed.

Etymology

The epithet refers to the microfruticose habit of the thallus consisting mostly of ascending isidioid structures.

Distribution and ecology

The species is known from several localities in south-west Florida (Lee Co.), where it inhabits the bark of Taxodium within and near the margins of seasonally flooded groves, at low elevation (c. 3–6 m).

Notes

Other isidiate species described in Porina appear to differ from this taxon in significant ways. Porina coralloidea P. James [=Zamenhofia coralloidea (P. James) Clauz. et Roux] has longer (40–57 μm), wider (9–13 μm) ascospores with considerably more numerous septa (9–11) and a very thick lateral wall (James 1971); its isidia are contorted chains of Trentepohlia surrounded by compacted hyphae (James 1971), rather than uniseriate structures surrounded by subglobose fungal cells. Porina rosei Sérusiaux has similar isidia corticated with globose cells, but the algal cells are not consistently uniseriate and the crustose thallus from which the isidia arise also has a cellular cortex of isodiametric fungal cells; its ascospores are wider (4–6(7) μm) and with only three septa (Sérusiaux 1991). Porina collina has fine fragile isidia, but they are not corticated nor is the phycobiont uniseriate. Porina isidioambigua M. Cáceres et al. has ascospores of about the same length as those of P. nanoarbuscula but twice as wide; the isidia incorporate many algal cells in width rather than a single file, and lack globose corticating mycobiont cells (Cáceres et al. 2013). Porina pseudohibernica has fine, abundantly branched, easily detached isidia, but their algal cells are not uniseriate, and their surface is ecorticate or covered with appressed hyphae rather than globose fungal cells; ascospores are longer (34–43 μm), wider (7–9 μm) and with more septa (7–8) compared to those of P. nanoarbuscula (Tretiach 2014). The mtSSU sequences furthermore indicate that P. nanoarbuscula and P. pseudohibernica are phylogenetically distinct (Fig. 9). The Floridian taxon P. scabrida (Harris 1995) is described as having isidia covered with a single layer of mycobiont cells, but its ascospores are longer (35–47 um), 8-celled, and almost twice as wide (5.5–7.5 um).

Additional specimens examined (all on Taxodium bark)

USA: Florida: Lee County, Fort Myers, Florida Gulf Coast University (FGCU) campus, Cypress grove along boardwalk to Parking Garage 3, 2020, W. B. Sanders 20423.1 (FLAS); ibid., Cypress swamp north of Parking Garage 3, 2021, W. B. Sanders 21320.1 (TSB 44456); ibid., along nature trail to laurel oak hammock, 2021, W. B. Sanders 21421.5 (FLAS, BR); ibid., along nature trail near laurel oak hammock, W. B. Sanders 21421.8 (FLAS); ibid., Cypress dome south of campus, near parking garage 2, 2020, W. B. Sanders 20511.3 (FLAS); ibid., 2021, W. B. Sanders 21502.2 (FLAS, BR); ibid., Cypress dome between FGCU Blvd and Aquatic Center, 2020, W. B. Sanders 20424.4 (FLAS); Estero, Grandezza, Cypress grove behind ‘The Studio’ sales centre, 2020, W. B. Sanders 20506.1 (TSB 44457); ibid., woods between Villa Grande and Grande Estates, 2020, W. B. Sanders 20510.2 (TSB 44458).

Discussion

Fruticose or isidiate crustose thallus?

Fruticose lichens, in the usual sense, bear their ascomata on their ascendant branches. In the Porina species examined here, perithecia develop on basal thallus portions embedded within the dead tissue layers of the plant substratum (Fig 1F & I–J, 2H). In this regard, their growth form corresponds to that of a crustose/endophloeodic lichen. Although true isidia are uncommon in crustose lichens, thalline structures of considerable anatomical, morphological, developmental and functional diversity are encompassed within this concept; previous authors have clearly found it broad enough to accommodate the appendages produced by certain species of Porina. However, in significant ways, the structural characteristics of the Porina species studied here call into question the utility of considering their ascendant structures as isidia. Isidia are supposed to be appendicular upgrowths of a corticated underlying thallus, such that there is continuity of the cortex and anatomy of the isidium with that of the thallus from which it arises (Jahns 1988). In our collections, the fruticose portions of the thallus predominate, while the basal crust is often evanescent, reduced, or largely integrated within the substratum. More significantly, the anatomical complexity of the corticated and/or stratified ascendant structures is nowhere evident in the diffuse, loosely organized substratic thallus from which they arise. The organized algal layer and medulla of P. microcoralloides and P. cf. scabrida, the corticating hyphae of the latter and the corticating layer of sub-globose fungal cells in P. nanoarbuscula are observed only in the ascending, isidia-like branches, without any counterpart in the unstratified basal portions. They therefore cannot be easily categorized as mere orthotropic outgrowths of the underlying crustose or endophloeodic thallus. We employ the term isidioid to describe the form of the ascendant vegetative thallus in the taxa described here.

Diversity of isidiate/isidioid taxa in Porina

Although isidia are not common in Porina, a number of species have now been described with such structures in this large genus. Our study suggests the presence of substantial genetic diversity among such taxa in south-west Florida, and amplifies previous indications that structures treated as isidia in this genus can be very different from each other anatomically. Other recent studies of Porina species have likewise found considerable genetic diversity (Orange et al. 2020), with many new taxa in the tropics (Ertz & Diederich 2022), including isidiate forms. In contrast, a number of isidiate collections worldwide, including from tropical Asia and South America, have been attributed to Porina hibernica (e.g. Aptroot 2003; Aptroot et al. 2017), a taxon originally described from Great Britain (Swinscow 1962). The seemingly cosmopolitan distribution of this species is thus noteworthy, but molecular data from such collections are needed to ensure that multiple, cryptic taxa are not involved. Structural and mtSSU sequence data (Fig. 9) from the present study indicated that none of the south-western Florida taxa examined correspond to P. hibernica.

Our sequence data provide further support to previous molecular studies (Orange et al. 2020; Ertz & Diederich 2022) that suggest multiple independent origins of isidiate morphology within Porina, showing this trait to be of little or no value as an indicator of biosystematic relationships. Strikingly, we found that the two isidioid taxa, P. microcoralloides and P. nanoarbuscula, appeared as sister taxa in the mtSSU analysis, yet their ascendant structures are so different in morphology and anatomical organization that one must suppose they originated independently of each other. On the other hand, the isidioid structures in P. microcoralloides were rather similar anatomically to those of P. cf. scabrida, the latter differing only in the presence of a better developed cortical layer. Yet these two taxa appear quite distant from each other in their mtSSU sequences. In at least some taxa, the presence or absence of isidia may also be variable. Ertz & Diederich (2022) found that isidiate morphs in certain species were scarcely different in nucleotide sequence from non-isidiate ones. In the case of our Floridian collections, where most of the vegetative thallus consists of the isidioid structures, it would be difficult to imagine conspecific morphotypes that lacked them. The extensive elaboration of the ascendant thallus component and concomitant reduction of the crustose portion suggests a trend by which fruticose thalli may evolve from crustose ones. Indeed, a transient basal crust is reported to precede the development of conventional fruticose structures in lichens such as Usnea (Lallemant 1984). While the biosystematic utility of isidia in Porina may be low, there is as yet no indication that more than one type of isidium or isidioid structure could occur within the very same taxon. It therefore seems reasonable to tentatively assume that details of their structure, at least when expressed, are characteristic of the taxon that possesses them, even if the trait tells us little or nothing about relationships to other taxa.

Also included within the concept of isidia is a distinctive structure produced by certain other Porina species (formerly Phyllophiale) that colonize leaves in tropical forests, forming symbioses with the multicellular discoid phycobiont Phycopeltis (Santesson 1952; Lücking & Cáceres 1999; Lücking 2008). The disc-shaped propagules are positioned on a very short central stalk above the thallus, from which they are easily detached. This type of isidium develops when a phycobiont filament (or filaments) from the margin turns upwards to emerge from the lichen surface and then branches radially in a plane parallel to the thallus below, accompanied by investing mycobiont hyphae on its upper and lower surfaces. The miniature lichenized disc perched above the main thallus can resume growth directly after detachment and dispersal (Sanders 2002). These isidia are unbranched, determinate structures (at least prior to detachment) with a highly uniform morphology likely adapted to water dispersal; they have little in common with the fruticose structures studied here. It has been pointed out that virtually identical discoid isidia are produced by foliicolous taxa of Porina representing different phylogenetic clades (Lücking & Vězda 1998; Lücking & Cáceres 1999), an interpretation supported by the positions of two such taxa (Porina alba (R. Sant.) Lücking and P. fusca Lücking) in gene-based cladograms (Orange et al. 2020; Ertz & Diederich 2022; see also Fig. 9). Thus, the discoid and the coralloid ‘isidia’ known in Porina share at least one feature: both appear to have arisen more than once independently. Although unreliable as indicators of biosystematic relatedness, such remarkable convergences do tell us something biologically important: that natural selection in these cases is almost certainly shaping morphologies with real and direct functional significance to the organisms involved.

Symbiont interactions

The consistent association of Porina microcoralloides and P. nanoarbuscula, each with a distinct clade of Trentepohlia phycobionts (Figs 11; Supplementary Material Fig. S2, available online), suggests a high degree of selectivity in these lichen-forming fungi. The correspondence of sister-clade pairs in the mycobiont and phycobiont trees suggests a possible occurrence of parallel cladogenesis in these symbiont lineages. It is noteworthy that both taxa occur in the same habitats and are even intermixed (Fig. 1N), where they are likely to encounter, and presumably reject, the algal strain preferred by the other species. A much larger dataset would be needed, however, before one may be confident that this is consistently the case.

It has been asserted that specialized penetrative contacts generally do not occur between symbionts in trentepohliaceous lichens (Nienburg 1926; Grube & Lücking 2002). However, this viewpoint was challenged by Tschermak (1941), who reported and illustrated deeply penetrating haustoria in numerous taxa of such lichens. Later TEM studies documented extensive haustorial development in almost every trentepohliaceous lichen examined (Withrow & Ahmadjian 1983; Matthews et al. 1989; Tucker et al. 1991; but see Lambright & Tucker 1980). These haustoria appear to finely invaginate the algal cell wall rather than actually traverse it. They would thus be classified as intraparietal (Type 2) in the scheme presented by Honegger (1986), although they often reach deeply into the algal cell (Tschermak 1941; Matthews et al. 1989). The haustoria observed in the Porina lichens examined here were all intraparietal, but with very limited intrusion causing only slight invagination of the algal cell wall (Figs 2G, 5E, 7H & 8E), as in Type 1 of Honegger (1986). They thereby contrast with the much deeper penetration/invagination of algal symbionts observed in the more typically crustose trentepohliaceous lichens as cited above. Why stratified and morphologically more complex foliose and fruticose lichens should have more superficial symbiont contacts than unstratified crustose lichens is unclear. One possibility is that mycobiont growth must be more closely coordinated with algal cell division to achieve more complex levels of organization (Greenhalgh & Anglesea 1979; Honegger 1987b), requiring more superficial attachments that can stimulate but not obstruct division and distribute its products. Another notable feature of the symbiont contact zones observed here is the very substantial thinning of the fungal cell wall at the point of its intrusion into the algal wall. Although lichen haustoria are not believed to play any central role in transfer of carbon, which is leaked by the alga across its walls in symbiosis, the reduction in mycobiont wall thickness does suggest a modification that serves to streamline exchange of substances. Symbiont contact zones of this type will be explored in more detail in a future work.