Carlquist’s growth ring classification: a functional approach that reinforces porous and annual rings

Summary – There is no doubt that Carlquist is one of the greatest contemporary wood anatomists. Every student or researcher in the field is likely to have read at least one of his numerous works, and it is difficult to find a topic in wood anatomy that does not lead to one to his publications. Among the many topics addressed by Carlquist are of course growth rings, one of the oldest and most exciting topics in wood anatomy. Carlquist established a functional classification for growth rings based on anatomical variation, in addition to an interpretation of the different anatomical markers “as ecologically adaptive devices.” In this article, I return to Carlquist’s first classification of growth rings and summarize the main changes in his later reviews. I highlight the advances provided by his unprecedented functional approach to growth rings, while also criticizing Carlquist’s proposed growth ring classification, which reinforces a partial view, although dominant until the present day, focused on temperate species: that growth rings are mostly well marked, porous or semi-porous, and with annual periodicity.

plants.Without information from these taxonomic groups, some of which were exclusively researched by Carlquist, a comparative approach as broad and robust as that conducted by Silva et al. (2021) would not have been possible.
Among the many topics addressed by Carlquist, it is no surprise that they include growth rings, one of the oldest and most exciting topics in wood anatomy (Glock 1941;Studhalter 1955;Studhalter et al. 1963).Observations, citations, and attempts to understand growth rings date back to the establishment of the scientific fields of wood anatomy and dendrochronology, which progressively developed from the second half of the 17th century onwards (Grew 1682;Malpighi 1686;Baas 1982).The diversity and functions of these structures aroused Carlquist's interest in the early stages of his career as a wood anatomist, and in 1980 he presented a first classification of growth rings accompanied by functional interpretations based on the variations found in the anatomical structure of the secondary xylem (Carlquist 1980).This classification, which offers considerable anatomical and functional detail, was revised, and published in two editions of his book Comparative Wood Anatomy (Carlquist 1988(Carlquist , 2001)).A summary of the growth ring type, subtype (when present), functions, and general occurrence by Carlquist classification, following the latest update on Carlquist (2001), is shown in Table 1, as well as diagrams that illustrate the types and subtypes presented (Figs 1 and 2).Carlquist was one of the pioneers of this field, since he was the first scientist to attempt a wide-ranging account of the functional anatomy of the xylem and presented a detailed description of the structure of several growth ring types, proposing functional hypotheses for these variations (Carlquist 2012).
By the time of the first classification proposed by Carlquist (1980), a considerable amount of knowledge had already been produced about growth rings (e.g., Coster 1927Coster , 1928;;Frost 1930aFrost ,b, 1931;;Kribs 1935Kribs , 1937;;Stallings et al. 1937;Chowdhury 1939Chowdhury , 1964;;Reed & Glock 1939;Glock & Reed 1940;Glock 1941Glock , 1951;;Studhalter,1955;Studhalter et al. 1963;Fritts 1966), especially in dendrochronology, which is concerned with the correlation between growth ring width and environmental variables, almost exclusively in subtropical and temperate tree species (Schweingruber 1996).Although other anatomical markers have been described, different from those traditionally recognized by dendrochronologists and wood anatomists of temperate and subtropical regions (Coster 1927(Coster , 1928;;Dechamps 1973;Fahn et al. 1981), little attention was paid to them.Likewise, discussions about the role of anatomical variations occurring in the secondary xylem responsible for the growth rings recognition (which we will refer to as "anatomical markers", although Carlquist did not use this terminology) were quite limited (Silva et al. 2019).Anatomical markers can be defined as the histological contrast that emerges from the traits that vary between cells produced throughout a growing season, forming early-and latewood: diameter, wall thickness, frequency of a given cell type, or even a change to another cell type.Each of these variations, isolated or combined, produce recognizable marks in the secondary xylem, which are repeated throughout the growing seasons (Silva et al. 2019(Silva et al. , 2021)).
In this article, I return to Carlquist's first classification of growth rings (Carlquist 1980) and summarize the main changes in his later reviews (Carlquist 1988(Carlquist , 2001)).I highlight the advances provided by his unprecedented functional approach to growth rings, while also criticizing Carlquist's proposed growth ring classification, which reinforces a partial view, although dominant until the present day, focused on temperate species: that growth rings are mostly well marked, porous or semi-porous, and with annual periodicity.

Carlquist's first classification of growth rings
Growth ring types were first categorized by Carlquist in the article Further concepts in ecological wood anatomy, with comments on recent work in wood anatomy and evolution, published in 1980 in Aliso: A Journal of Systematic and Evolutionary Botany.In this article, he established a functional classification for growth rings based in the combination of different anatomical variations he considered important, largely focusing on variation in vessel diameter and frequency, in addition to interpreting this anatomical diversity "as ecologically adaptive devices" (Carlquist 1980, p. 499).Carlquist emphasized the incomplete nature of his classification (Carlquist 1980), which was The majority of ring-porous dicotyledons fall into this category.
Type 4: Vessels wider in earlywood, but instead of grading into narrower latewood vessels, there is a rather marked change in vessel diameter from earlywood to latewood.
[. . .] essentially the same as with type 3, but with a brief period of peak water conduction each season, followed by a long period during which conduction is slower and latewood conductive safety is of value.
Mediterranean-type climates, in which a brief rainy season is followed by a prolonged dry season.Type 12: Wood diffuse-porous, but with an annually produced parenchyma band.
[. . .] seasonal storage of starch related to flushes of growth.
Many diffuse porous dicotyledons have annual flushes of growth (e.g., many Araliaceae and Fabaceae).Type 13: Vessels markedly dimorphic between earlywood and latewood: vessels wide in diameter and not in extensive groups in earlywood; latewood vessels narrow and formed in extensive groupings that traverse rays and may be oriented either diagonally (intermediate between tangential and radial) or tangentially.
[. . .] wide earlywood vessels that would accommodate peak flow.However, the latewood appears to be an unusually well-developed series of mechanisms for conductive safety: either very narrow vessels or (very commonly) vasicentric tracheids are present in the latewood vessel bands, and thus elements that embolize more tardily (very narrow vessels or vasicentric tracheids) are intermixed with the narrow latewood vessels.Additionally, the diagonal and tangential bands have the effect of forming a network, perhaps interconnecting virtually all vessels within a growth ring, which assures maintenance of the conductive pathways regardless of embolism formation in a large number of vessels.
-Type 14: Vessels are markedly wider in earlywood, and the wood is thus ring-porous.The growth ring is delimited by markedly thicker walled imperforate tracheary elements.
The functional significance [. . .] does not appear to be one of enhanced safety in the latewood fiber-tracheids.
-Type 15: Vessels grade from wider in earlywood to narrower in latewood, although the difference in diameter is not as marked as in typically ring-porous woods; imperforate tracheary elements are either absent or present, only in latewood, as a pair of fiber strands in each fascicular area.
revised, featuring deletions and new inclusions of growth ring types, in the first edition of his 1988 book Comparative Wood Anatomy, with a revised second edition published in 2001 (Carlquist 1988(Carlquist , 2001)).
In this first growth ring categorization, 13 growth ring types were proposed, four of which have subtypes, with a total of 15 subtypes.Type I is distinct from the other types, being considered less defined when compared to other types, visually with little contrast between early-to latewood, or in Carlquist's words, "little differentiation into growth rings; changes with season gradual" (Carlquist 1980, p. 503), described as "moderate differentiation into growth rings ('semi-ring porous' if vessels present); changes with seasons gradual" in later revisions (Carlquist 1988(Carlquist , p. 17, 2001, p. 14), p. 14).It is important to point out that this characterization of the Type I growth rings as poorly defined by Carlquist, is made in comparison to porous rings.Type I when compared with tropical tree growth rings, mainly recognized by the anatomical marker thick-walled and/or radially flattened latewood fibers, they would be regarded as well defined (Silva et al. 2019).Type II represents a well-marked ring type, restricted to woods with the exclusive presence of tracheids as a tracheary element, typical of temperate gymnosperms, although this type also occurs in  (Carlquist 1988(Carlquist , 2001) ) with the following caption: "Fig.2.1.Diagrams of the growth ring types.Conventions are at top, left.Subtypes (A etc.) are shown for some of the types.For further explanation, see text.All of the growth ring types shown here are also illustrated by one or more photomicrographs in the remaining photographs of this chapter" (Carlquist 2001, p. 12).Reproduction authorized by Springer.

6
IAWA Journal Downloaded from Brill.com 10/07/2023 05:15:11AM via Open Access.This is an open access article distributed under the terms of the CC BY 4.0 license.https://creativecommons.org/licenses/by/4.0/some angiosperm taxa in which vessels are absent, particularly in taxa distributed in temperate zones.Types III-XI are to porous or semi-porous rings.
The growth ring types described in Carlquist's classification combine several criteria, such as the degree of distinctness of rings, variation in diameter, thickness, and frequency in the different types of cells (mainly regarding vessel elements and tracheids), as well as the imperforate tracheary element types.In my view most of these types can be categorized using the three traditional anatomical markers considered by dendrochronologists and wood anatomists in temperate regions.These markers are: (i) thick-walled and/or radially flattened latewood tracheids (types IA and II), (ii) semi-ring-porosity (types IB-IE) and (iii) ring-porosity (types III-XI) (Reed & Glock 1939;Glock & Reed 1940;Glock 1941Glock , 1951;;Studhalter 1955;Studhalter et al. 1963;Fritts 1966;Schweingruber 2007).Of the 13 types and their variations, only two do not include the three anatomical markers mentioned above.These two types are: "Type XII.Wood diffuse porous but initial parenchyma present" and "Type XIII.The occurrence of cork bands (interxylary cork)" (Carlquist 1980, p. 523).Even so, Type XII is mentioned only as a possibility.It was initially cited by Chowdhury (1964), but not formalized by Carlquist, who was unsure whether it could be compared to the other growth rings considered in his classification -thus, no functional hypothesis is proposed for Type XII.Type XIII, on the other hand, was regarded as a rare phenomenon found mainly in certain desert shrubs, and may function in "channeling water in particular portions of the xylem" (Carlquist 1980, p. 525).
In his paper, Carlquist presented descriptions of the anatomical identification of growth ring types and subtypes, as well as photomicrographs (42 photos distributed in 10 plates), functional hypotheses, and associations with climatic types and/or taxonomic lineages for most types and subtypes (Carlquist 1980).

Revisions of Carlquist's growth ring classifications in the first and second editions of Comparative Wood Anatomy
periodicity is only made for the marker regarded as uncommon for temperate trees; (iii) the exclusion of "Type XIII.The occurrence of cork bands (interxylary cork) demarcating growth rings" (Carlquist 1980, p. 523).This exclusion is justified given that interxylary cork does not indicate growth ring formation, because in the genera in which it occurs, such as Artemisia, growth rings are similar to each other across species and are recognized regardless of the presence or absence of interxylary cork (Carlquist 1988); (iv) three new growth ring types are proposed -Types 13, 14, and 15 (Table 1) -all of which are porous or semi-porous; although Type 14 seems to recognize the role of latewood fibers in growth ring demarcation, these are only recognized for porous ring conditions; (v) two plates with 25 diagrams were incorporated, depicting all types and subtypes of the growth rings described by Carlquist (Figs 1 and 2).These types and subtypes were also illustrated in one or more of the 50 photomicrographs that make up the 12 plates across the chapter on growth rings (Carlquist 1988(Carlquist , 2001)).

A pioneering approach of the growth ring function
One of the great merits of Carlquist's perspective lies in its pioneering approach to growth ring functionality, especially considering the anatomical changes observed between early-and latewood.The general approach prevailing at the time of Carlquist's first publications (1980,1988), which still prevails today, largely ignores the different anatomical markers and their functional significance in discussions about growth rings.That is, the discussions are focused on environmental seasonality or the environmental factors that would modulate growth ring formation, disregarding the different anatomical markers of this analysis.(e.g.Alves & Angyalossy-Alfonso 2000; Nath et al. 2016;Arévalo et al. 2017).However, if there are differences between species in growth ring anatomical markers, the functional solutions might also be distinct (Carlquist 2012).The current approach of simply correlating growth rings with environmental variation without detailed study of the anatomical markers and functional relevance of these correlations is almost certainly because research with growth rings is led by dendrochronology, which seeks to identify and quantify environmental events through the dating of annual growth rings (Babst et al. 2017;Islam et al. 2018;Quesada-Román et al. 2022).Much of dendrochronology involves searching for correlations between the widths of growth rings and environmental variables (Schweingruber 1996(Schweingruber , 2007;;Evert 2006).This search for ring widthclimate correlations allows for the construction of chronologies and has several applications, such as in climatology, ecology, geomorphology, and archaeology (Jacoby 1989;Stahle et al. 1999;Roig 2000;Tomazello Filho 2009;Morales et al. 2020;Kuitems et al. 2022) but ignores the functional discussion about anatomical markers.In fact, most of the dendrochronology literature can be confused in this regard, for example with notions that the environmental variables can "cause" or "control" wood anatomy, rather than anatomy being a product of natural selection.
Carlquist mostly discussed the function of growth rings from the perspective of the putative trade-off between efficiency in earlywood, with wider vessels/tracheids, and the safety conferred by latewood, with narrower and/or thicker walls vessels/tracheids, or with the exclusive presence of tracheids or those associated with the narrower vessels in the latewood (Table 1).The mechanical stiffness function of latewood, as a way of compensating for the low resistance of earlywood is also considered, although with less emphasis (Baas et al. 2004).This functional approach of Carlquist could not be different in view of the great emphasis given to porous anatomical marker in your growth rings classification, even when other anatomical markers are considered, such as marginal parenchyma or thickwalled and/or radially flattened latewood fibers (Table 1).Although at the time of the first publications (Carlquist 1980(Carlquist , 1988)), a consensus had already been established that wider vessels in earlywood increase conduction rates, the presence of narrower vessels in latewood remains poorly understood.Thus, Carlquist hypothesized that narrow vessels were safer, with a lower risk of embolism formation (Carlquist 1975(Carlquist , 2022) ) and were thus a positively selected characteristic in plants that periodically suffer high stresses in the xylem, either by cold or by drought (Hacke et al. 2001(Hacke et al. , 2017;;Venturas et al. 2017).This hypothesis would subsequently gain support thanks to favorable evidence  (Carlquist 1988(Carlquist , 2001) ) with the following caption: "Fig.2.1.Diagrams of the growth ring types.Conventions are at top, left.Subtypes (A etc.) are shown for some of the types.For further explanation, see text.All of the growth ring types shown here are also illustrated by one or more photomicrographs in the remaining photographs of this chapter" (Carlquist 2001, p. 13).Reproduction authorized by Springer.obtained through comparative and experimental studies (Tyree & Zimmermann 2002;Sperry et al. 2008;Hacke et al. 2017;Venturas et al. 2017;Islam et al. 2018;Olson et al. 2018;Savage et al. 2022).
Carlquist's functional perspective on wood anatomy can be attributed to his extensive knowledge on this subject, developed over decades of research, through careful observation, collection, and systematization of anatomical, morphological, and environmental data of the taxa he worked with.This allowed Carlquist to interpret secondary xylem anatomical patterns in light of "ecological" anatomy, that is, the study of wood features as adaptations (Carlquist 2001(Carlquist , 2022)).The research strategies developed by Carlquist are very similar to those employed by the English naturalist Charles Darwin (1808-1882), these research strategies are considered fundamental in the proposition of the theory of evolution in 1859 (Ghiselin 1969;Hull 1973).Carlquist wrote on his science communication website that the comparative method provided ideas for new experiments and hypothesis testing.The observation of wood under a microscope is to Carlquist like seeing the results of a giant evolutionary experiment, where cell types and sizes indicate how the plant works (Carlquist 2022).
The anatomical patterns described by Carlquist associated with different floras and climates are fundamental to understanding how natural selection can produce wood anatomy diversity.These anatomical patterns described by Carlquist and their association with environmental variables and the plant life form (trees, shrubs, succulents, etc.) allowed the elaboration of the fundamentals of the functional wood anatomy.Several functional hypotheses and assertive conclusions, later corroborated in comparative and experimental studies, were postulated by Carlquist.I present an example precisely about the porous rings, of how the seasonal production of the secondary xylem adjusts in a synchronized way to the climatic variations of the seasons: "A growth ring contains within it contrasting modes of adaptation to different climates, as well as the climatic events of a particular year.Earlywood of a particular species may be likened to wood of a tropical rain forest tree in which wide vessels accommodate large volumes per unit time to suit active transpiration; such wood is low in conductive safety.Latewood may resemble wood of a desert shrub, in which large numbers of narrow vessels offer greater resistance to embolism formation and thus greater conductive safety (Hargrave et al. 1994) and redundancy in case some of them do embolize."(Carlquist 2001, p. 11).

Carlquist's growth ring classification: overemphasis on porous and annual rings
Despite its great contributions to growth ring research, especially regarding function, some criticisms can be directed toward Carlquist's classification.His classification strongly reinforces the traditional approach adopted by dendrochronologists and wood anatomists of temperate regions, which prevailed until recently, in which growth rings are recognized only when they are essentially annual, strongly demarcated, and restricted to the most pronounced markers.As a result, this leaves out many tropical species whose growth rings do not fit into this restricted view (Studhalter et al. 1963;Silva et al. 2019).On this topic, it is important to emphasize that tropical regions show a richness of tree species much higher than subtropical and temperate regions (Slik et al. 2015).This temperate-centric notion of growth rings ignores growth rings with more weakly defined anatomical markers, such as thick-walled and/or radially flattened latewood fibers, which occur widely in the world's flora, including many tropical species with diffuse porous wood, or even those that have infra-annual (multiple annual rings) or supra-annual rings, or rings without regular periodicity (Silva et al. 2019(Silva et al. , 2021)).Based on the small diversity of anatomical phenomena of wood in the temperate zone, Carlquist's growth ring classification remains excessively narrow in scope.
Despite the importance of the notion, Carlquist's works never define "growth ring."In fact, he claims that there are no criteria to define whether growth rings are present or absent in the secondary xylem: "Unlike Chowdhury (1964), I do not have criteria whereby growth rings are said to be present or counted as not present" (Carlquist 1980, p. 503).A general and less subjective concept, based on anatomical markers that are globally distributed and not just from the temperate zone, was recently proposed by Silva et al. (2019).I see that the greatest contribution of the classification of growth rings proposed by Carlquist is in the functional discussion and not in the typology itself.Although the classification was produced with the purpose of developing hypotheses regarding growth-ring function (Carlquist 1980), most types and subtypes described show convergence for the discussion about safety and hydraulic efficiency (Table 1).That these many types are putatively functionally similar leads us to question whether a less extensive classification would not be more appropriate.Carlquist made his descriptions of the different types of growth rings by comparing the latewood of a ring with the earlywood of the subsequent ring.Although this procedure in principle shouldn't introduce major discrepancies with general practice, it does differ from the general current procedure for studying growth rings, in which the anatomical markers are defined by comparing the early-and latewood within the same growth ring (Silva et al. 2019(Silva et al. , 2021)).
Of all classifications of growth rings, I am aware of (e.g., Coster 1927Coster , 1928;;Glock 1941;Studhalter et al. 1963;Dechamps 1973;Fahn et al. 1981;IAWA Committee 1989;Worbes 1989Worbes , 1995Worbes , 2002;;Worbes & Fichtler 2010;Nath et al. 2016;Silva et al. 2019), Carlquist's is the most extensive (Carlquist 2001), although he leaves out some anatomical markers, as we will see below.Analyzing Carlquist growth ring classification, I was unable to identify a dominant criterion used by him.Carlquist's growth rings classification is based on the combination of several traits that emerge from different criteria.These criteria include growth ring distinctness, justifying the separation of Type 1 from the others; diameter, wall thickness and frequency of different cell types in the secondary xylem; presence/absence of vascular and vasicentric tracheids; perforation plate type of vessel elements; imperforate tracheary element types; change in cell type, conductive and/or not; and in vessel arrangement/grouping type between early-and latewood; it also included rather subjective criteria such as the degree of variation in diameter and frequency of the vessels, using descriptors such as "moderately wide", "moderate differentiation", "much wider", "moderately wide" and "difference in diameter is not as marked".His inclusion of so many subjective and different criteria led to an extensive classification with 15 types and 14 subtypes, in some cases of very restricted distribution, such as Types 9 and 10, or hard to differentiate, e.g., Types 5 and 8 (Carlquist 2001).
Despite this large number of types and subtypes, almost all the functional hypotheses converge on the tradeoff between hydraulic safety and efficiency (Baas et al. 2004).In practically all of Carlquist's functional hypotheses regarding growth rings, earlywood features are less resistant to embolism but more conductively efficient, whereas latewood features are more resistant to embolism but less conductively efficient.This similarity of functional scenarios is perhaps not surprising given the relatively restricted anatomical and climatic range of variation on which Carlquist based his classification.Most of his types and subtypes are based on small variations of porous, semi-porous and thick-walled and/or radially flattened latewood tracheid layers rings (Table 1).
In Carlquist's classification (Carlquist 1980(Carlquist , 1988(Carlquist , 2001)), he highlighted the unfinished nature of his classification, which was based on the anatomical variants known to him.Carlquist ignored, whether intentionally or not, publications prior to his which already made proposals for classifying growth rings (e.g., Coster 1927Coster , 1928;;Glock 1941;Dechamps 1973;Fahn et al. 1981;IAWA Committee 1989).In Carlquist's first classification (Carlquist 1980) all the 13 growth ring types he proposed, including the subtypes (except for only two types, one of which is abandoned in subsequent publications), had thick-walled and/or radially flattened latewood tracheids, semi-ring-porous or ringporous (Silva et al. 2019(Silva et al. , 2021)).At the beginning of the 20th century, it was demonstrated that growth rings could be demarcated by other anatomical markers, based on tropical species, in addition to those traditionally recognized for temperate and subtropical trees, such as thick-walled and/or radially flattened latewood fibers, marginal parenchyma, and fiber zones.These anatomical markers were highlighted in later publications focusing on tropical species (Jane 1934;Chowdhury 1936Chowdhury , 1953Chowdhury , 1964;;Fahn et al. 1981;Worbes 1989Worbes , 1995Worbes , 2002;;Marcati et al. 2006;Worbes & Fichtler 2010;Wilczek et al. 2014) and by the IAWA Committee (1989).
Carlquist's emphasis on porous and annual growth rings can be seen not only in the definitions of the types in their classification (Table 1) but is also reinforced at several points throughout his writings (Carlquist 1980(Carlquist , 1988(Carlquist , 2001)).When Carlquist claimed that diffuse-porous wood was omitted from his classification for the sake of definition, he clearly assumes that there are no growth rings in wood with this type of porosity: "Notable in this regard is the transition between diffuse porous (by definition omitted from the classification below) and semi-ring porous" Downloaded from Brill.com 10/07/2023 05:15:11AM via Open Access.This is an open access article distributed under the terms of the CC BY 4.0 license.https://creativecommons.org/licenses/by/4.0/(Carlquist 2001, p. 14).This emphasis is made clear when Carlquist presents a synthesis of families with growth rings at the end of the chapter in the topic "Systematic Distribution of Growth Rings" when only general types "markedly ringporous," "ring-porous at most but commonly semi-ring-porous," and "semi-ring-porous or ring-porous in a minority of species" are presented (Carlquist 2001, pp. 36-37).
Carlquist also states that all the growth rings presented in his works exhibit annual periodicity, although he does not clarify how this periodicity is tested, or whether it is simply inferred by the strong anatomical marking and seasonality of the sampled environments (Carlquist 1980(Carlquist , 1988(Carlquist , 2001)).He rejects the formation of successive cambia in the growth ring definition on the grounds of a lack of annual or seasonal periodicity: "The formation of successive cambia, which often do not coincide with annual or seasonal phenomena, are not treated in this classification" (Carlquist 1980, p. 501).Another example of emphasis on annual periodicity can be seen in the example below on marginal parenchyma.Carlquist recognizes the possibility of the existence of growth rings with periodicities other than non-annual and emphasizes the importance of identifying these different variations: "One may be grateful for studies which delineate clearly which woods in a group or flora lack growth rings, which woods tend to have a growth ring of a nature other than annual, and which woods have annual growth rings" (Carlquist 1980, p. 501).
Marginal parenchyma is shown in the first Carlquist (1980) classification, but only when it is associated with semiporous and porous rings (Type IC and Type XI)."Type XII.Wood diffuse-porous but initial parenchyma present" (Carlquist 1980, p. 523)," is mentioned as only a possibility by Carlquist, referring to this anatomical marker that was described at the time by Chowdhury (1964).However, Carlquist is unsure whether it represents growth ring activity in the sense of the other types described in his work, probably with respect to annual periodicity.In the two revisions that followed this first classification, "Type 12. Wood diffuse-porous, but with an annually produced parenchyma band" (Carlquist 1988, p. 33;2001, pp. 29-30) (Table 1), is recognized as a valid type, implying that this is more than a theoretical possibility.However, it is important to note that in the new classification, the annual periodicity is emphasized in the description of Type 12, demonstrating Carlquist's commitment to this restrictive view on the definition of growth rings: if there is any possibility that diffuse-porous wood presents a growth ring, this must be annual, like all other types presented in his classification (Carlquist 1980(Carlquist , 1988(Carlquist , 2001)).In his text he agrees with Chowdhury (1964), "that parenchyma bands formed in a nonannual fashion cannot be cited in connection with growth rings" (Carlquist 1988(Carlquist , p. 30, 2001, p. 26), p. 26).
Growth rings demarcated by thick-walled and/or radially flattened latewood fibers, the most common anatomical marker among flowering plants, abundant not only in tropical species (Silva et al. 2021), but also in subtropical and temperate species, are not formally considered in Carlquist's first classification (Carlquist 1980).Although he commonly describes the presence of tracheids, fiber-tracheids, and/or libriform fibers as imperforate tracheary elements for each growth ring type, variations in the diameter and/or wall thickness of libriform fibers and fibertracheids are reported only for Types ID, IE and VII (Carlquist 1980).For the types considered by Carlquist to be semi-ring-porous or ring-porous, it can also be assumed that these ring types also have the anatomical markers thickwalled and/or radially flattened latewood fibers.Although the thick-walled and/or radially flattened latewood fibers marker is largely associated with porous and semi-porous rings, these markers are rarely cited by wood anatomists in temperate regions (Silva et al. 2021).In his two subsequent revisions (Carlquist 1988(Carlquist , 2001)), Types 1D,1E, 14 and 15 all were primarily described as be semi-ring-porous or ring-porous, and so follow the temperate zone tradition of focusing mostly on vessels or tracheids.However, ancillary features did include thick-walled and/or radially flattened latewood fibers.As is already consensus, a growth ring can be recognized by one or more anatomical markers (IAWA Committee 1989;Silva et al. 2019).This fibers variation assumes secondary importance in Carlquist's growth ring types.In Type 14, greater emphasis is placed on fiber wall thickening than in the other types, although he didn't offer a functional hypothesis for this type.Variation in fibers alone has not been formally proposed in any of the types described by Carlquist, only considered when accompanied by porous or semi-porous rings.
Following the functional anatomy approach of studying secondary xylem developed by Carlquist, functional (and developmental) hypotheses were recently proposed to understand the evolutionary origin and retention of thick- walled and/or radially flattened latewood fibers, the most widely distributed anatomical marker in flowering plants.This marker is present in 75% of angiosperm species that have growth rings, and its high frequency in many species is almost certainly associated with functional advantages (Silva et al. 2021).We recently proposed three (non-mutually exclusive) hypotheses regarding development and function of thick-walled and/or radially flattened latewood fibers anatomical marker.The first hypothesis concerns alteration in the conditions favorable for growth and development in the vascular cambium can produce smaller and/or more sclerified cells, characterizing the latewood.The second was that thick-walled and/or radially flattened latewood fibers increase stem mechanical support and enable the formation of increasingly wide vessel elements across secondary xylem.The third is that imperforate tracheary elements with narrower lumina and thicker walls increase the safety of transport in vessels (Silva et al. 2021).When referring to angiosperm semi-porous-ring and porous-ring, Carlquist (1988Carlquist ( , 2001) ) stated that the anatomical variation in these growth rings reflects contrasting models of adaptation to different environmental demands.Here we propose that latewood fibers also act in the prevention of embolism in diffuse porosity wood.If this functional hypothesis is true, as indicated by recent research (Franke et al. 2002;Cochard et al. 2004;Brodribb & Holbrook 2005;Jacobsen et al. 2005), the variation found in growth rings demarcated by thick-walled and/or radially flattened latewood fibers are almost certainly adaptations to variation in precipitation and water availability, mainly in tropical wood species with diffuse porosity (Silva et al. 2021).
Two other anatomical markers were not included in the Carlquist's classification: fiber zones and distended rays.Fiber zones were mentioned in previous works (e.g., Coster 1927Coster , 1928;;Fahn et al. 1981;IAWA Committee 1989;Worbes 1989Worbes , 1995)), and distended rays appear in the IAWA Committee (1989) list.However, these anatomical markers are not treated from a functional perspective in any of these works.In a recent review of the phylogenetic and functional perspective of growth rings (Silva et al. 2021), we indicated that fiber zones likely have a similar function to that of thick-walled and/or radially flattened latewood fibers, in which the seasonal production of dense and mechanically resistant tissue may result from a physiological condition or may have been selected because it offers less risk of embolism formation in response to stresses in the xylem.Fiber zones in latewood can result from variations in different cell types and attributes within the secondary xylem: wall thickness/diameter of the fibers, diameter/frequency of the vessels, diameter/frequency of the cells of the axial parenchyma, absence of vessels and axial parenchyma, and wider or narrower spacing of the lines/bands of parenchyma.Assuming that these fibrous zone variations originated from distinct ancestral conditions in different lineages of angiosperms we presented four subtypes of fibrous zones according to the main variations found: (i) fiber zone with thick-walled fibers; (ii) fiber zone with axial parenchyma cells and/or vessels present at lower density and/or smaller in diameter, or even absent in some cases; (iii) wider spacing between parenchyma lines/bands; and (iv) narrower spacing between parenchyma lines/bands (Silva et al. 2021).
Distended rays are the most indistinct and uncommon of the anatomical markers described in the literature, although their commonness is likely to be underestimated.Distended ray function is not related to environmental variation, as in other anatomical markers.Instead, their development seems to be related to a structural issue by adjusting to the adjacent cells and preventing the formation of large intercellular spaces and consequently of cracks in the wood (Silva et al. 2021), similar to rays in the secondary phloem (Esau 1965;Evert 2006), and parenchyma tissues are widely known for their filling function (Fahn 1974;Dickison 2000;Evert 2006;Cutler et al. 2008).Distended rays can also be found in the wood not associated with growth rings, such as in the scar tissue caused by injuries (Molina et al. 2016), in parenchyma-like tangential bands (Silva et al. 2015), and in some species with scalariform or reticulate axial parenchyma (Fedalto et al. 1989).
There is no doubt regarding the important legacy of Carlquist regarding growth rings and many other topics in wood anatomy.A sampling of these contributions can be seen in articles in this Special Issue of IAWA Journal.Carlquist felt that studies of the functional anatomy of wood should reflect the overall proportionality of structural diversity found across plants globally (M. Olson, personal communication).However, most anatomical, and physiological studies of secondary xylem are concentrated in temperate and subtropical regions, in contrast to the greater plant diversity in Downloaded from Brill.com 10/07/2023 05:15:11AM via Open Access.This is an open access article distributed under the terms of the CC BY 4.0 license.https://creativecommons.org/licenses/by/4.0/tropical areas (Lamanna et al. 2017).Extrapolating the question asked by Studhalter (1955, p. 36), "What would the botany of today be if all early centers of botanical research and most of the later ones had been located in the tropics instead of in the Temperate Zone?," we may ask ourselves what the field of wood anatomy, and specifically, growth rings research would look like if the same number of studies were carried out in tropical regions.Undoubtedly, this field of research would look very different.
The growth rings approach proposed by Carlquist, as described throughout this paper, illustrates how, in biological research, the use of restrictive models limits broader discussions.This temperate-centric notion of growth rings, which was developed until the first half of the 20th century, focuses on porous, semi-ring, and well-marked rings, and with annual periodicity, as are commonly found in temperate tree species, demonstrates the way that certain characters with relatively low frequency within a given taxon (seed plants, in this case) can be taken as anatomicalfunctional models considered to be widespread and dominant.With gains being made in the dendrochronology and wood anatomy study of tropical species (e.g., Coster 1927Coster , 1928;;Worbes 1989Worbes , 1995Worbes , 2002;;Worbes & Fichtler 2010;Sonsin et al. 2013;Brienen et al. 2016;Nath et al. 2016;Pagotto et al. 2017;Schöngart et al. 2017;Silva et al. 2017Silva et al. , 2019Silva et al. , 2021;;Islam et al. 2018), it has gradually become clearer that flowering plants very often do have growth rings, much more frequently than previously predicted (Wheeler & Baas 1991;Wheeler et al. 2007).However, growth rings in tropical species are often not marked by ring-porous, semi-ring-porous, or latewood tracheids, but mainly by thick-walled and/or radially flattened latewood fibers, in addition to other anatomical markers, such as marginal parenchyma and fiber zones (Silva et al. 2021).We still do not know how environmental variations, especially water availability, affect the functioning and survival of tropical plants, nor the role played by different anatomical markers in this context, with the same understanding that we have about temperate species (Silva et al. 2021;Olson 2022).
In our search for a global functional wood anatomy, specifically with regards to growth rings, there is a need to expand our sampling practice and include the structural diversity (hence, functional diversity) that exists in woody tissues.In particular, special attention must be paid to tropical plants, typically found in threatened regions, and which at times are treated as exotic or anomalous (e.g., Gasson & Dobbins 1991;Dickison 2000;Carlquist 2001;Heklau et al. 2012), compared to the more restricted, but better known anatomical and functional models found in temperate species.Thus, it is these species should have a more prominent place within the contemporary context of plant anatomy and function.

Fig. 1 .
Fig. 1.Carlquist growth ring types and subtypes diagrams: types 1A-5D.These plates were published in the first and second editions of the Comparative wood anatomy book(Carlquist 1988(Carlquist , 2001) ) with the following caption: "Fig.2.1.Diagrams of the growth ring types.Conventions are at top, left.Subtypes (A etc.) are shown for some of the types.For further explanation, see text.All of the growth ring types shown here are also illustrated by one or more photomicrographs in the remaining photographs of this chapter"(Carlquist 2001, p. 12).Reproduction authorized by Springer.

Fig. 2 .
Fig. 2.Carlquist growth ring types and subtypes diagrams: types 6-15.These plates were published in the first and second editions of the Comparative wood anatomy book(Carlquist 1988(Carlquist , 2001) ) with the following caption: "Fig.2.1.Diagrams of the growth ring types.Conventions are at top, left.Subtypes (A etc.) are shown for some of the types.For further explanation, see text.All of the growth ring types shown here are also illustrated by one or more photomicrographs in the remaining photographs of this chapter"(Carlquist 2001, p. 13).Reproduction authorized by Springer.

Table 1 .
Summary of the growth rings type, subtype (when present), functions, and general occurrence by Carlquist classifications.