ECOLOGICAL FACTORS IN WOOD EVOLUTION: A FLORISTIC APPROACH
SHERWIN CARLQUIST
Reprinted from AMERICAN JOURNAL OF BOTANY
Vol. 64, No. 7, August 1977 Made in the United States of America
Amer. J. Bot. 6 4 ( 7 ) : 887-896. 1977.
ECOLOGICAL FACTORS IN WOOD EVOLUTION: A FLORISTIC APPROACH1
SHERWIN CARLQUIST Claremont Graduate School, Pomona College, and Rancho Santa Ana Botanic Garden.
Claremont, California 91711
A B S T R A C T
Wood florulas from southwestern Australia were analyzed to determine whether wood anatomy is sufficiently correlated with ecology so that vessel element features can be said to have a predictive value. Indices for vulnerability (vessel diam: vessels per sq. mm) and meso-morphy (vulnerability x vessel element length) were calculated for each species in the following florulas: karri forest understory, coastal granitic slopes, bogs, sand heaths, and desert. Wood indices for the species studied and for each florula show that these florulas form a sequence in increasing xeromorphy in the order listed. Genera represented in more than one florula validate the trends. Data for Gyrostemonaceae. Loranthaceae, and Cupressaceae are calculated separately because these are succulents, epiparasites, and conifers, respectively. Comparison with categories from floras elsewhere in the world shows the flora of Western Australia as a whole to be relatively xeromorphic. The indices devised show promise of great reliability because correlations with rainfall, temperature, and other factors are very close. Functional nature of the vessel element is thereby believed to be clarified.
IN ATTEMPTING to analyze factors underlying the selective forces which have guided evolution of conductive tissue in vascular plants, I have used various types of correlation (1975a). This method could be criticized as constituting indirect evidence. The most direct type of evidence would appear to be experimental. Experimental work, such as that initiated by Scholander et al. (1975), represents valid lines leading to a potential synthesis between wood anatomy and physiological ecology. Ideally, one would like to isolate physiological factors and compare them to individual wood characteristics; however, wood cannot be isolated from the many adaptations of the plant in which it exists or from the complex ecological regimes to which any given species is adapted in its tolerance ranges. Foliar apparatus, for example, can be of overriding importance. High diffusive resistance of leaves and consequent lowered transpiration rate, crassulacean acid metabolism, or C4 photosynthesis are among the factors that can override xylem conformtion as an optimal design system for coordinating foliage with conductive efficiency and resistance to tension of water columns in tracheary elements.
Experimental work on woods, if it is to be done meaningfully, must be structured on hypotheses.
1 Received for publication 13 December 1976; revision accepted 6 April 1977.
This study was aided by grants from the National Science Foundation, GB-38901 and BMS 73-08055 A-l . The author is indebted to Dr. Larry DeBuhr for assistance in collection of data. Mr. A. S. George and Mr. Kevin Richards kindly invited me on an expedition far into the interior of Western Australia.
Correlations based upon wood anatomy and habitat of given species seem a valid approach to production of these hypotheses. One may proceed to develop correlations either with a systematic group or with a floristic unit. The former method was used, for example, where comparisons between wood anatomy and species, species groups, and their ecology and habit were offered (Carlquist, 1975b, in press). Onagraceae cover a wide range of habit from annual to arboreal, and range from tropical cloud forest to dry areas and aquatic habitats, although Onagraceae include no true desert shrubs or lianas. The almost perfect habital and ecological correlations with wood anatomy in that family prompted extension in the use of quantitative vessel element features to such families as Penaea-ceae (Carlquist and DeBuhr, in press). Penaea-ceae occupy a relatively wide range of habitats for a small family, but not as great as Onagraceae. This fact is reflected in quantitative terms with regard to wood anatomy.
My concepts concerning the latter two families are an extension of my (1975a) hypothesis, and are as follows. Short, narrow vessel elements are theorized to resist high tensions in water columns. Narrowness of vessels is, to some extent, inversely correlated to number of vessels per sq. mm. This inverse correlation is, however, by no means a perfect one. By dividing mean vessel diameter by number of vessels per sq. mm of transection, one finds a range of values, not a constant, within a family (Carlquist, in press; Carlquist and DeBuhr, in press). A low value for this ratio could be interpreted as great "redun-
887
8KS AMERICAN JOURNAL OF BOTANY [Vol. 64
dancy" of vessels. The more numerous the vessels per sq. mm. the less the chance that disabling of a given number of vessels by air embolisms formed under water stress would seriously impair conduction in a plant. This idea was suggested to me by Dr. Martin H. Zimmermann (personal communication). A low value for this ratio would therefore indicate capability of withstanding water stress or freezing. I have termed this ratio "vulnerability" (Carlquist, in press). That ratio, multiplied by mean vessel element length, I termed "mesomorphy" in accordance with the demonstration that higher values occur in species that seem, on various grounds, more highly mesomorphic.
The rationale underlying use of vessel element length is that short vessel elements are more resistant to collapse or deformation than long ones because of the strengthening furnished by the end walls, which are unpitted constrictions along the length of a vessel. Vessel element length, furthermore, is controlled by morphogenetic factors independent of those affecting diameter of vessels and number of vessels per sq. mm of transection. Other subsidiary characters which may be related to xeromorphy could be added: thickness of vessel walls, number of bars per perforation plate, and the nature of lateral wall pitting of vessels. However, the majority of dicotyledons have simple perforation plates. The thickness of vessel walls and the tendency for vessels to be grouped in large aggregations or to be formed in a solitary fashion are characters stabilized in particular taxonomic groups rather than susceptible to a wide range of modification in dicotyledonous families at large. This is also true of lateral wall pitting of vessels. Although these subsidiary features are worthy of consideration, the three vessel element measurements cited as utilized in the indices devised were chosen because they are applicable to virtually all dicotyledonous woods.
The alternative to study of a taxon in the ecological interpretation of wood structure is, as mentioned, a floristic one. One can trace this approach to such authors as Webber (1936), who studied wood of chaparral and desert shrubs. More recent representatives of the floristic approach include Novruzova ( 1968) and Versteegh ( 1968). Novruzova's samples did not include a broad ecological range, and only some of the desired quantitative data were collected. Versteegh did not deal quantitatively with wood features and used altitudinal levels (with ecology only implied) as a criterion.
Both the floristic and systematic approaches are complementary, as implied by the work of Baas (1973). One might question the validity of a floristic approach if, as 1 allege (1975a, p. 30-31). each species has its own series of anatomical solutions in coping with water relations.
and if species with different xylem formulations— even excluding vascular plants other than dicotyledons—can coexist in literally the same locality. If wood characteristics of particular flor-ulas tend to conform to particular plans and to differ modally from those of other florulas, however, we have a tool, just as in the systematic approach, for developing criteria for xylem adaptation to varying ecological regimes. Even if these tools fall short of utmost precision, we can use clear correlations in developing a holistic approach to expressing water relation adaptations of individual species, incorporating foliar and I other features in addition to wood anatomy.
Ideally, one would wish for both systematic and floristic comparisons within a single study. ^ In the present study, as many genera and families as possible which occur in more than one florula were selected: the genera Acacia, Boro-nia, Casuarina, Hibbertia, Hybanthus, and Pime-lea; the families Epacridaceae, Fabaceae, Myrta-ceae, Proteaceae, Rutaceae, and Sterculiaceae. for example. In a sense, I used both approaches in my 1966 summary of wood anatomy of Astera-ceae, but the ecological and habital categories were relatively simplistic, and florulas were not used.
MATERIALS AND METHODS—During my 1974 field work in Western Australia. I obtained wood samples from several distinctive habitats. Western Australia contains well-marked ecological zones and a flora that is characterized by considerable adaptive radiation into these zones (Carlquist, 1974). These considerations are basic to my choice of southwestern Australia, but no single region is ideal. Ultimately a variety of different regions should be compared with each other. The comparisons offered by the "World flora" categories in Table 6, however, provide an equally valid kind of perspective.
Wood samples of desert shrubs in Western Australia were obtained from a transect beginning at Cocklebiddy on the Nullarbor Plain coast, ranging northwards through the Victoria Desert and the Gibson Desert to the Rawlinson Range near the southwestern corner of Northern Terri- « tory. The remaining ecological zones sampled r were relatively small in extent by comparison. The karri (Eucalyptus diversicolor) forest under-story shrubs were collected in the vicinity of Pem-bsrton and Manjimup. Coastal shrubs were all collected at a single station, Canal Rocks, where they grew in deep pockets of soil on granitic domes. Bog shrubs were collected in wet depressions between Albany and Nornalup. The sand heath shrubs sampled were obtained from the large sandplain inland from the Jurien Bay coast near Badgingarra. The Western Australian sand heath is composed of aeolian acidic sands. The bogs present a problem with regard to their eco-
Au«USt, 1977] CARLQU 1ST—FACTORS IN WOOD EVOLUTION
TABLE 1. Karri forest understory shrubs (Annual rainfall, ca. 150 cm)
889
Vessel Vessel Vessels element diam persq. length,
Species Family Coll. no. „.:n mm nm V M
Acacia urophvlla Benth . F a b a c e a e 5563 57 31 262 1.84 4 8 2
Acacia s p . F a b a c e a e 5576 51 39 280 1.31 367 Bossiaea laidlawiaiui F a b a c e a e 5571 61 48 217 1.27 550
T o v c y & M o r r i s Casuarina decussata Ben th . C a s u a r i n a c e a e 5569 42 71 433 0.59 255 Chorilaena quercifolia End l . R u t a c e a e 557(1 53 100 389 0.53 206 Hibbertia furfuracea Dil len iaceae 6065 37 95 642 0.39 250
( R . B r . ) Benth . H. tetrandra ( L i n d l . ) Gi lg Di l l en iaceae 5581 34 86 743 0.36 267 Hoxea elliptica ( S m . ) D C . F a b a c e a e 5582 42 74 249 0.57 142 Lasiopetalum floribundum Ben th . S te rcu l iaceae 6 0 7 4 30 145 279 0.21 59 Leucopogon verlicillalus R. Br. Epac r idaceae 5 5 6 4 32 98 460 0.33 152 Pimelea clavata Labi l l . T h y m e l a e a c e a e 5565 65 45 190 1.44 274 Trvmalium spathulatum R h a m n a c e a e 5577 48 57 474 0.82 389
(Lab i l l . ) Ostf.
F l o r u l a a v e r a g e 4 6 74 385 0.62 239
logical status, a problem discussed in a later section.
The rainfall figures in Tables 1-5 are from the Atlas of Australian Resources (1952-1966). In Table 6, the figures for Gyrostemonaceae are based on 15 collections representing three genera and seven species. An account of wood anatomy of this family will be published elsewhere. The Loranthaceae of Table 6 represent four collections of Loranthus (sensu lato), parasitic on acacias, from the desert transect. The conifers included in Table 6 are Actinostrobus acumina-tus, A. pyramidalis, Callitris canescens, and C. robusta. Collection numbers in the tables are my own, and voucher specimens are located in the herbarium of the Rancho Santa Ana Botanic Garden. The figures on Arctic shrubs in Table 6 are compiled from data on eight species in a wood
florula from Greenland published by Miller (1975). The remaining "World flora" data in Table 6 are based upon Carlquist (1975a).
Woods were sectioned and macerated according to the usual techniques. The means of the lengths and diameters of vessels elements (tra-cheids in the case of conifers) were obtained from 50 measurements per feature in each species. The mean number of vessels per sq. mm of transection was obtained from 10 measurements for each collection. The reader may question the omission of such statistical details as standard deviation and standard error—items which could easily have been calculated from the raw data in hand. However, I find that such statistical analysis is unwarranted in the present study because there are overriding sources of potential and accidentia! bias. Standard deviation and standard
TABLE 2. Coastal shrubs from Canal Rocks (Annual rainfall ca. 100 cm)
Species Family Coll. no.
Vessel diam lira
Vessels persq.
mm
Vessel element length,
urn V \ i
Acacia sp. F a b a c e a e 6083 49 61 224 0.80 179 Boronia alata S m . Ru taceae 5541 4 2 163 358 0.26 93 Diplolaena dampieri Desf. R u t a c e a e 5544 32 146 525 0.22 116 Exocarpus odorafus ( M i q . ) D C . San ta l aceae 5545 4 8 102 182 0.47 8 6 Exocarpus sparteus R .Br . San ta l aceae s.n. 28 299 501 0.09 45 Guichenotia ledifolia J . G a y . S te rcu l iaceae 5 5 4 6 28 196 2 5 4 0 .14 3 6 Hibbertia cuneifortnis Dil len iaceae 6085 41 71 755 0.58 4 3 8
(Lab i l l . ) Gi lg Hibiscus huegelii E n d l . M a l v a c e a e 6088 63 35 3 1 6 1.80 568 Pimelea sp . T h y m e l a e a c e a e 5 4 5 4 4(1 107 196 0.37 73 Ricinocarpus sp . E u p h o r b i a c e a e 6087 43 7 6 318 0.57 181 Scaevola crassifolia Labi l l . G o o d e n i a c e a e 6084 37 119 368 0.31 114 Templetonia retusa ( V e n t . ) R.Br . F a b a c e a e 6086 4ii 43 187 0.93 174
Florula average 41 118 342 0.35 120
890 AMERICAN JOURNAL OF BOTANY [Vol. 64
TABLE 3. Bog shrubs (Annual rainfall ca. 150 cm)
Vessel Vessel Vessels element cliam persq.
Species Family Coll. no. um mm fem V M
Acacia mooreana W. V. Fitz. F a b a c e a e 6047 61 35 2 4 9 1.76 4 3 8 Boronia denticulata Sm. R u t a c e a e 5 7 7 0 31 I 11 226 0.28 63 Boronia oxvantha Turcz. R u t a c e a e 5705 23 182 253 0.12 30 Cosmelia rubra R.Br. E p a c r i d a c e a e 5675 25 765 587 0.03 IS lsopogon buxifolius R.Br. P r o t e a c e a e 5720 25 1 6? 3 0 4 0.15 4 6 Leptospermum crassipes Lehm. M y r t a c e a e 5724 64 140 333 0.49 163 Leptospermum firmum M y r t a c e a e 5708 35 207 559 0.17 95
(Schau.) Benth. Leucopogon assimilis R.Br. E p a c r i d a c e a e 5707 25 3 2 4 4 8 2 0 .08 39 Melaleuca micropltxlla Sm. M y r t a c e a e 5712 4:; 175 4 9 8 0.24 119 Melaleuca sp. M y r t a c e a e 6007 37 37 331 1.00 331 Persoonia longifolia R.Br. P r o t e a c e a e 6058 36 120 530 0.30 159 Phebalium argenteum Sm. Rutaceae 5 7 1 0 3 ! 209 3 1 4 0 .15 47 Pultenaea dasvphvlla Turcz. F a b a c e a e 5 7 0 6 41 121 210 0 .34 71 Stirlingia tenuifolia P r o t e a c e a e 5711 33 144 183 0.23 42
( R . B r . ) S teud .
F l o r u l a a v e r a g e 37 195 361 0.19 69
error indicate statistical validity for cell populations for which figures were actually obtained. They cannot tell us anything about materials not studied, however. For example, only one stem per species was studied. Other portions of a given plant and other individuals of a given species would have given different figures that would render meaningless standard deviation and standard error on the single stem studied, judging
from the results of Stern and Greene (1958). The selection of species was dictated by routes of travel and places where stops were made, hardly the ideal sampling criteria. The ecological limits of a florula are a matter of arbitrary decisions, and thus a source of human bias. Since each species has a wood formulation that works conjunctively with a particular type of foliar apparatus, one could say that an attempt should have been
TABLE 4. Saiul heath shrubs (Annual rainfall ca. 65 cm)
Species Family Coll. no.
Vessel diam um
Vessels per sq.
Vessel element length,
um M
Cyanostegia lanceolata Turcz. Cyanostegia sp. Eremaeopsis pauciflora (Endl.) Druce Eriostemon spicaius A. Rich. Geleznowia verrucosa Turcz. Halgania lavandulacea Endl. Hibbertia lineala Steud. Hybanthus bilobus C. A. Gardn. H. floribundus (Walp.) F. Muell. Lachnostachys eriobotrya
(F. Muell.) Druce Leucopogon australis R. Br. Lysinema ciliaium R. Br. Physopis lachnostachya C. A. Gardn. Pileanthus peduncularis Endl. Pimelea sp. Pityrodia bartlingii
(Lehm.) Benth. P. old field ii (F. Muell.) Benth. Rhagodia preisii Moq. Santalum spicatum (R.Br.) DC. Xylomelum angustifolium Kippist.
Florula average
Dic ra s ty l idaceae 5905 36 22 1 274 0.16 44
Dic ras tv l idaceae s.n. = 6 252 2 5 7 0.14 36 M y r t a c e a e 5 9 4 4 36 180 323 0 .20 65 R u t a c e a e 5 9 4 5 24 4 6 2 385 0 .05 19 R u t a c e a e 5895 23 3 0 0 192 0 .08 15 Borag inaceae 5744 21 614 263 0.03 S Di l len iaceae 5940 34 217 605 0 .16 97 V io l aceae 5748 23 482 302 0 .05 15 Vio laceae 5298 2') 246 5 1 2 0.12 61
Dic ras ty l idaceae 5896 34 155 266 0.22 5')
E p a c r i d a c e a e 5549 32 209 4 2 7 0 .15 '-4 Epac r idaceae 5942 is 522 340 0.04 14
Dic ras ty l idaceae 5975 36 109 201 0.33 66
M y r t a c e a e 5890 27 139 313 0.20 61 T h y m e l a e a c e a e 5939 34 183 276 0.19 52
Dic ras ty l idaceae 5483 44 171 251 0 .26 55
Dic ras tv l idaceae 5899 36 93 210 0.39 S2 C h e n o p o d i a c e a e 5440 37 133 95 0.28 27
San ta l aceae 5911 48 70 271 0.68 185 P r o t e a c e a e 5889 57 41' 367 1.41 518
33 240 307 0.14 43
August, 1977] CARLQUIST—FACTORS
TABLE 5. Desert shrubs (Annual rainfall ca. 22 cm)
Specie Family
Acacia burkitlii (F. Muell.) Benth. Fabaceae A. sowdenii Maiden Fabaceae A. tetragonophylla F. Muell. Fabaceae A. sp. Fabaceae A. sp. Fabaceae A triplex nitmmularia Lindl. Chenopodiaceae Casuarina pinaster C. A. Gardn. Casuarinaceae Eremophila latrobei F. Muell. Myoporaceae Halgania cyattea Lindl. Boraginaceae H. sp. Boraginaceae Hvbamhus auianiiacus Violaceae
(F. Muell.) Mclch. Kochia sedifolia F. Muell. Chenopodiaceae Myoporum sp. Myoporaceae Myoporum sp. Myoporaceae Prostanthera sp. Lamiaceae Sollya sp. Pittosporaceae
Florula average
made, for example, to select species representing a range of foliar types within each florula. Epa-cridaceae have exceptionally narrow vessels, quite numerous per sq. mm of transection. Had a larger number of Epacridaccae been selected for any of the florulas, a greater degree of xero-morphy for that florula would have appeared. The difficulties of sampling when a floristic approach is used thus render sophisticated statistical analysis superfluous. The means and indices in the tables, however, can be said to be valid when interpreted cautiously.
From Table 6, one sees that a "V" value of approximately 1.0 to 2.5 indicates mesomorphy, whereas a figure below 1.0 indicates redundancy of vessels and greater safety under conditions of water stress. Values for "M" would indicate what one would call mesophytes above levels of approximately 200, if one also concedes that succulents may be called mesophytes in terms of xy-lem structure. Xcrophytes in the classicial sense would have values of 75 or below.
RESULTS—The florulas and other groups can be reported in terms of the categories for the Western Australian flora listed in Table 6.
Karri forest understory shrubs—The values shown in Table 1 indicate that understory shrubs of the karri forest are easily the most mesophytic dicotyledons in the Western Australian flora. This accords not only with the greater annual rainfall of the karri forest areas, but also with the finding by Gindel (1973) that despite transpiration in a wet forest, soil moisture remains higher than in open areas. However, the karri understory shrubs are not notably mesophytic
WOOD EVOLUTION 89 J
Coll. no.
Vessel diam .vm
Vessels per sq.
mm
Vevel element length,
/im V M
5136 43 109 215 0.39 84 5140 (.1 57 167 1.06 177 5139 43 70 214 0.61 131 5137 54 57 163 0.95 154 5145 55 92 231 0.60 139 5141 28 99 91 0.29 26 5980 28 226 406 0.13 53 5184 29 321 163 0.09 L5 5167 37 136 192 0.27 52 5156 27 298 175 0.09 16 5170 33 328 295 0.10 29
5142 28 144 83 0.19 16 5134 36 125 234 0.29 67
5143 34 165 221 0.21 46 5160 27 426 196 0.06 12 5276 28 413 430 0.07 30
37 192 217 0.19 41
when compared with the woods of primitive dicotyledons (Table 6). The criteria for selection of those primitive woods were: more than 10 bars (mean) per perforation plate of vessel elements; presence of true tracheids as the imperforate tra-chcary element; and presence of diffuse axial parenchyma. Of the karri understory shrubs, only three have values of less than 200, suggested above as a lower threshold for mesomorphy: Hovea elliptica, Lasiopetalum floribundum, and Leucopogon verticillatus. Of these, Hovea elliptica and Lasiopetalum floribundum could be described as not true karri understory shrubs, for they appear most commonly along road margins in relatively exposed areas. Leucopogon verticillatus is more typically an understory plant, but may appear along road cuts as well. Leucopogon appears to share with all Western Australian Epacridaccae a very xeromorphic wood formulation, however, and one can hypothesize that phylads of Epacridaceae evolving into more mesic situations do not alter wood formulations nearly so much as foliar apparatus (Leucopogon verticillatus has broad, thin leaves unlike the needle-like leaves of most Leucopogon species).
Coastal shrubs—As Table 2 shows, V and M values for woods in this ecological category average about half the values for the karri understory shrubs. This can be seen in terms of individual genera (Acacia, Pirnelea) as well as in the trend of the figures. This habitat can be regarded as intermediate between karri understory and sand heath in terms of rainfall, available water in the form of runoff from granitic domes at Canal Rocks, and the mitigating influence of maritime humidity in modifiying temperature and transpiration extremes.
892 AMERICAN JOURNAL OF BOTANY [Vol. 64
TABLE 6. Wood anatomy of Western Australian flortilas compared to categories from other areas
V» ssel diam, Vessels Vessel element World flora /tin per sq. m m length, um V M
Mesic primitive woods 109 47 1385 2.29 3172 Rosette trees 79 31 412 2.25 1051 Vines and lianas 157 19 334 8.22 2745 Annuals 61 162 1X6 0.38 71 Desert shrubs 29 353 218 0.08 17 Stem succulents 72 64 259 1.33 344 Arctic shrubs V 559 245 0.10 25
Western Australian Flora Karri understory shrubs 46 74 385 0.62 239 Coastal shrubs 41 118 349 0.34 119 Bog shrubs 37 195 361 0.19 69 Sand heath shrubs J 2 240 307 0.14 43 Desert shrubs 37 192 217 0.19 41 Gyrostemonaceae 71 62 180 1.15 206 Loranthaceae 36 155 78 0.23 18 Conifers' (27) i 1596. (0.02)
' Figures for conifers are for tracheids rather than vessel elements.
Bog shrubs—These woods (Table 3) were collected in order to determine whether shrubs respond in evolutionary patterns to what would seem, at first glance, a highly mesic habitat. One must note that although designated as "bogs'" in Australian floristic literature, these habitats do not correspond to bogs in the sense of floristic literature of the North Temperate Zone. The Australian bogs, like the "vleis" of Cape Province. South Africa, are not comparable to bog habitats or other ecological categories described by northern hemisphere botanists. The Australian bogs are depressions with underlying hard-pan which hold standing water during the winter and early spring months, unlike depressions in sand heath country. Bogs in Western Australia vary in depth and duration of standing water. As Table 3 shows, only one species. Acacia moor-eana. appears to show mesomorphic response to this habitat. The remainder of the species can be said to show remarkably low V and M values. The explanation for this appears twofold. First, the bog areas of southwestern Australia are very few and small in extent compared to dryland areas, and are therefore unlikely to have been centers of preservation of a relictual mesic flora (one notable exception might be Cephalotus fol-licularis). The phytogeographic source for bog species has very likely been the sand heath flora, which is composed of species with similar apparent preference for acidity. The bogs, like the sand heaths, have a substrate of white granitic sand, so this ecological shift would be a logical one. One must remember, moreover, that even bogs in this moderately high rainfall area are subjected to considerable heat and drought from November through April. For example, in the study area where bog species were collected, average rainfall falls below 7 cm per month from Novem
ber through March (Atlas of Australian Resources, 1952-1966). During the extremes of these months, these bogs are actually equivalent to sand heaths in the dryness of their porous, sandy bottoms. The bogs of southwestern Australia are mesic only in having a longer period of water availability than do sand heath areas. Obviously, extremes rather than averages in rainfall and temperature influence wood structure. The bog species can be said to differ from those of sand heath areas in their tolerance to inundation (the wood structure of Leptospermum eras-si pes suggests this) during the relatively brief rainy season. The xeromorphic structure of wood in bog shrubs has a positive selective value for drought, but this structure is not of selective disadvantage during periods of water availability.
Sand heath shrubs—The V and M values for bog shrubs (Table 3) are, as might be expected, slightly higher than for sand heath shrubs (Table 4) . Although in a belt of rainfall with more than twice the rainfall of the desert shrubs (Table 5) , the sand heaths are, in essence, a sandy desert during the summer months. The water-holding capacity of sand heaths is certainly minimal. Summer extremes provide a water relations requirement regime equivalent to a desert, a fact indicated by the near identity of V and M values for sand heath shrubs and for desert shrubs. That the sand heath shrubs represent more xeromorphic wood structure than do the coastal shrubs can be seen by comparisons within individual genera such as Hibbertia, Pimelea, and Eriostemon (a close relative of Boronia). The vulnerability values of sand heath shrubs are remarkably low compared with the "World flora" figures of Table 6. Lower values can be found in dicotyledon assemblages only in the desert
August, 1977] CARLQU 1ST—FACTORS IN WOOD EVOLUTION 893
shrubs (from North America) and Arctic shrubs (from Greenland).
Desert shrubs—To be sure, the source of the desert shrubs (Table 5) used in this study is a long transect rather than a single area. The transect consists of (1 ) shallow clay-like lateritic soils overlying limestone on the Nullarbor Plain; (2) red sands of the Victoria Desert; (3) stony lateritic clays and sands of the Gibson Desert. The annual rainfall of these areas is so low (see Table 5) that water availability of clay versus sand would seem to be of little importance. Clay soils might permit the growth of small trees, such as some species of Acacia, however. All of the acacias represented in Table 5 came from the Nullarbor Plain. The Nullarbor Plain has slightly higher rainfall and possibly greater water availability (clay soils above a limestone hardpan) than do desert areas to the interior. In fact, if one substracts the acacias from Table 5. one obtains much lower values for the desert assemblage (vessel diameter = 30 ^m; vessels per sq. mm = 244; vessel element length = 226 /im; V = 0.12; M = 27). These V and M values are sufficiently lower than those of the sand heath shrubs so that the lower rainfall and extreme heat of the Victoria and Gibson Deserts appear reflected in wood data.
Noteworthy with respect to the desert shrubs is that the two with successive cambia (Atriplex nummularia and Kochia sedifolia) conform in quantitative features to woods of dicotyledons with normal cambia. In the sampling of woods with successive cambia in my earlier (1975a) study, the greater width of parenchyma bands lowered the number of vessels per sq. mm, and thus these woods appear more mesomorphic quantitatively than do the Australian desert woods with successive cambia. In the desert Atriplex and Kochia, parenchyma bands between products of successive cambia are so reduced that the woods agree in all quantitative details with woods of desert shrubs with normal cambia. The same can be said for sand heath shrubs of the family Dicrastylidaceae. All species of Dicrastylidaceae have successive cambia, but quantitative data for these woods fall within the range of woods of sand heath dicotyledons with normal cambia. Epacridaceae are virtually absent from desert regions, probably because interior sands are alkaline, whereas the sands of the sand heaths are acidic, a condition preferred by Epacridaceae.
Gyrostemonaceae—As noted in the previous section, removal of the Nullarbor Plain acacias gives a more accurate picture of desert wood characteristics. The Gyrostemonaceae were deliberately removed from the remainder of the flora not only because their quantitative features differ markedly from those of other shrubs in
their respective habitats (chiefly sand heath, with some species in desert and coastal sands), but also because examination of bark and leaves of Gyrostemonaceae shows them to be succulents. Succulents are rare in the Australian flora at large. One might regard the underground tubers so prevalent in genera such as Drosera, Chamae-scilla, etc. as forms of succulence, however. The presumption that Gyrostemonaceae qualify as succulents seems justified. If one compares figures for Gyrostemonaceae (Table 6) to those of "stem succulents" from the World flora at large (Table 6) , one notes a virtual identity in quantitative features.
Loranthaceae—The Western Australian Lo-ranthaceae show a very low M value, as one would expect, since epiparasites would be expected to have wood more xeromorphic than that of their host plants (Carlquist, 1975a). This xeromorphy takes a form different from that of desert shrubs, as the data on Loranthaceae in Table 6 show. The Loranthaceae studied do not have an exceptionally low figure for "vulnerability." This would be related to the probability that mistletoes do not experience sharp seasonal fluctuations in moisture availability. Water in xylem of the epiparasite would be expected to be under very high tension, however. If so, the shortness of vessel elements in Loranthaceae would agree with my (1975a) hypothesis on resistance of short vessel elements to high tensions in water columns of the xylem.
Perforation plates—Table 7 shows the two dicotyledonous families in the flora of southwestern Australia in which scalariform perforation plates in vessels are known. To this list may be added Byblis gigantea (Byblidaceae), in which a bar or two traversing a perforation plate may be seen occasionally (Carlquist. 1976). The species within each family in Table 7 are arranged in order of decreasing number of bars per perforation plate. Undoubtedly other species of Hibbertia and of Epacridaceae from southwestern Australia have scalariform perforation plates. Epacridaceae, Byblis, and Hibbertia may be regarded as mesic elements that have become established in mesic pockets of southwestern Australia and radiated into drier habitats. The remainder of the flora of southwestern Australia is a "simple perforation plate flora." indicative of establishment by groups with simple perforation plates and otherwise more xeromorphic wood features (Carlquist, 1975a). Epacridaceae, despite having predominantly scalariform perforation plates, are not incongruous in a flora that can, as a whole, be called xerophytic. As the figures for Epacridaceae in Tables 1-6 show, they have very narrow vessels, numerous vessels per sq. mm, moderately short vessel elements, and therefore, as a family, have
894 AMERICAN JOURNAL OF BOTANY [Vol. 64
TABLE 7. Species with SCalariform perforation plates in the flora of southwestern Australia
Species Coll. no. Habitat Bar per
plate, mean
Dilleniaceae Hibbertia telrandra (Lindl.) Gilg H. cuneiformis (Labill.) Gilg H. furfuracea (R. Br.) Benth. H. lineata Sleud.
Epacridaceae Cosmelia rubra R. Br. Sphenotoina dracophylloides Sond. Andersonia echinocephala (Stschegl.) Druce Leucopogon assimilis R. Br. Lysinema ciliatum R. Br. Leucopogon verticillatus R. Br. Leucopogon australis R. Br. Leucopogon atherolepis Sischegl.
5581 6085 6056 5940
5675 5691 5692 5707 5942 5564 5549 5690
karri understory 25.3 coastal scrub 23.7 karri openings 9.5 sand heath 7.0
bogs 16.6 montane scrub 14.8 montane scrub 8.0 bogs 2.3 sand heath 1.1 karri understory 1.1 sand heath 0.4 montane scrub 0.0
remarkably low V and M indices. Consequendy, Epacridaceae may be said to be adapted (or "pre-adapted") to xeric situations. The bars on the perforation plates in Epacridaceae from southwestern Australia are not thin and wiry, as in such tropical mesophytes as Illicium; they are wide and bordered. The fewer the bars, the wider they are; also, the wider the perforations between the bars. The occurrence of wide, relatively few bars and large perforations in the generally xeric (compared with wet forest) habitats of the species in Table 7 could be cited in support of my hypothesis that few, wide bars with large perforations represent a modification for xeromorphy in a phylad with scalariform perforation plates (Carlquist, 1975a, p. 160). The figures of Table 7 and the wood of Byblis gigantea (Carlquist. 1976) suggest that groups with relatively primitive wood show accelerated specialization during adaptation to the generally dry climate of southwestern Australia.
Epacridaceae seem well suited in wood structure and foliage to xeric conditions in Western Australia. Only Leucopogon verticillatus, from the moist karri understory. has broad leaves, and these probably represent a secondary expansion of the lamina during evolutionary response to shady conditions. Although only a scattering of species of Leucopogon have been surveyed, this genus can be said to have predominantly simple perforation plates. The fact that Leucopogon is the largest genus of Epacridaceae in Western Australia and has radiated so successfully in sand heath areas is very likely associated with the reduced number of bars per perforation plate, in accordance with my (1975a) hypothesis regarding ecological distribution of scalariform versus simple perforation plates. An additional factor worthy of note is that the imperforate elements in the wood of Epacridaceae are all tracheids. Tracheids are, at least in theory, much better for
conduction than libriform fibers, which are present in woods of most dicotyledonous families. Tracheids are potentially much better for conduction than are libriform fibers because of greater pit membrane area. Thus, even if a very high proportion of vessels of Epacridaceae were blocked by air embolisms formed under water stress, the tracheids would represent a supplementary conductive system that could conduct an appreciable volume of water.
Precisely the same considerations as offered for Epacridaceae would hold true for Hibbertia as well. Unfortunately, my sample of sand heath species (which develop very small amounts of wood and therefore tend to be overlooked during wood collecting) consists of only a single species. Hibbertia lineata, however, has the fewest and widest bars per perforation plate and the widest perforations of any of the species studied. Perhaps other sand heath species of Hibbertia will prove to be similar, but additional study is highly desirable. With respect to woods of Epacridaceae and Hibbertia in Western Australia, one must also take into account the nature of the foliar apparatus. Many Western Australian species of Hibbertia have linear or ephemeral leaves. Those with moderately broad leaves may well have high diffusive resistance, which would limit transpiration just as effectively as reduced leaf area. Unfortunately, diffusive resistance of leaves has been measured for only a few angiosperm species, but would surely be an extremely significant measurement with regard to water relations and wood structure.
Other vessel features—The three quantitative features related to vessel measurements may be modified by other features. Greater vessel wall thickness can be an expression of xeromorphy, presumptively functioning in resistance to higher water tensions, as in Larrea (Carlquist, 1975a).
August, 1977] CARLQUIST—FACTORS IN WOOD EVOLUTION 895
In other groups, such as Asteraceae, larger groupings of vessels with increasing xeromorphy may occur (Carlquist, 1966). This would be a functional correlation explainable by the hypothesis that grouped vessels resist water tensions more effectively by mutual support than do isolated vessels (Carlquist, 1975a). Features such as these may well contribute to xeromorphy of woods in particular groups. If so, the vessel features of prime importance in Tables 1-6 may be expected to vary with relation to these modifying vessel features.
Conifers—Although not really comparable to dicotyledonous woods, conifers represent a xylem plan ideally adapted to water stress conditions. If "redundancy" of conducting cells in the wood is a measure of viability of xylem when air embolisms form, conifers are unexcelled. In addition, air embolisms, when they occur, can be localized within individual cells in the case of tra-cheids, but in dicotyledonous vessels elements, they can spread from one vessel element into an entire vessel. If one utilizes tracheids instead of vessels as a basis for calculating a "V" ratio, one can see that an extremely low vulnerability is demonstrated by conifers of Western Australia (Table 6) , which are probably not markedly different in this respect from other conifers, although tropical conifers (Agathis, Araucaria, Podocarpus) tend to have somewhat wider tracheids and therefore fewer tracheids per sq. mm of transection. Conifer xylem represents a conductive system successful in dry areas of Australia only if transpiration rates are low, as they presumably are in the microphyllous Cupressa-ceae of Western Australia. Although much lower in a "V" ratio than vessel-bearing dicotyledons, the lowest ratio in the latter group is in Arctic shrubs (Table 6) . This is not surprising, for water in xylem of these shrubs is undoubtedly completely frozen during the winter. Air bubbles are therefore present in all xylem elements when spring thawing occurs. Probably these Arctic shrubs have mechanisms for resorbing or otherwise expelling air from vessels, but great redundancy of conductive cells in secondary xylem is undoubtedly functionally valuable in this regime. Species with scalariform perforation plates and with simple perforation plates are represented in about equal bumbers in Miller's (1975) Greenland wood florula, so the nature of perforation plates is probably a minor consideration in conductive characteristics. There is, however, a large proportion of conifers and of dicotyledon woods with tracheids or fiber-tracheids as imperforate tracheary elements (Empetraceae, Eriaceae, Rosaceae) in Miller's florula. As mentioned in the preceding section, dicotyledonous woods with tracheids rather than libriform fibers as imperforate tracheary elements might be expected to
have greater "safety" under conditions of freezing as well as water stress. A high degree of redundancy is not necessarily primitive It can be found in such a highly specialized dicotyledon as Loricaria thuyoides, which has very narrow vessels and a large number of vascular tracheids —an almost conifer-like wood (Carlquist, 1975a).
CONCLUSIONS—Those skeptical of analysis of xylem function by means of anatomical correlations may point to the fact that within each of the florulas studied here, there is not strict conformity to a narrow quantitative range. The mitigating features of foliar structure (e.g., cuticle thickness; succulence) and physiology (crassulacean acid metabolism; C4 photosynthesis) are obvious sources for departure from a particular xylem formulation. Other factors, such as the other vessel features mentioned above and the nature of the root system (spread, depth, cork covering, presence of mycorrhizae) are additional reasons for departure from a uniform plan within a florula. Ranges within each florula are, however, of a relatively small order of magnitude compared with World flora categories (Table 6) . The averages derived from each florula suggest correlations between xylem anatomy and habitat. The values for these indices fall in precisely the same sequence as the probable mesomorphy or xeromorphy of the habitats. In turn, woods of all of the florulas of Western Australia conform to a relatively xeromorphic pattern, with V and M values at best well below those of primitive woods chiefly from tropical mesic regions. The usefulness of the V and M indices as ecological indicators seems well substantiated on the basis of systematically-oriented as well as floristically-oriented approaches (or a mixture of the approaches).
Byblis, Hibberita, and Epacridaceae may represent establishments of mesomorphic groups that have been able to radiate subsequently into drier environments. Among non-woody dicotyledons, one might add Cephalotus, Drosera, and Lentibulariaceae. The vast bulk of the dicotyledonous flora of Western Australia, however, has specialized xeromorphic woods. This majority of the flora may have established itself in moderately dry to dry areas of Western Australia, and radiated mostly into drier habitats but also, in an appreciable number of instances, into more mesic habitats such as the karri understory as well. Radiation into more mesic habitats by phylads appears to have occurred in other floras, such as that of the Hawaiian Islands (Carlquist, 1974, 1975a), where Asteraceae would be a prime example. We tend to think in terms of increasing adaptation to xeromorphy as a generalization— and this may be true in the majority of instances. However, this phylesis is probably reversible under special circumstances.
•
896 AMERICAN JOURNAL OF BOTANY [Vol. 64
LITERATURE CITED
A n AS oi- A I S I R M I A N RISOIRCTS. 1952-1966. Department of National Development. Canberra (parts published variously, many not credited to particular authors).
BAAS. P. 1973. The wood anatomical range in Ilex (Aquifoliaceae) and its ecological and phylogenetic significance. B lumea21 : 193-258.
CARLQUIST, S. 1966. Wood anatomy of Compositae: a summary, with comments on factors controlling wood evolution. Aliso 6(2 >: 25-44.
. 1974. Island biology. Columbia University Press, New York and London.
. 1975a. Ecological strategies of xylem evolution. University of California Press. Berkeley.
. 1975b. Wood anatomy of Onagraceae, with notes on alternative modes of photosynthate movement in dicotyledon woods. Ann. Mo. Bot. Gard. 62: 3X6-424.
. 1976. Wood anatomv of Byblidaceae. Bot. Gaz. 137: 35-38.
. 1977. Wood anatomy of Onagraceae: additional species and concepts. Ann. Mo. Bot. Card, (in press i
. AND L. DF.BUHR. 1977. Wood anatomy of Penaeaceae (Myrtales): comparative, phylogenetic. and ecological implications. Bot. J. Linn. Soc. (in press).
GINDEL. I. 1973. A new ecophysiological approach to forest-water relationships in arid climates. Dr. W. Junk B.V.. Publishers. The Hague.
MILLER. H. J. 1975. Anatomical characteristics of some woody plants of the Angmagssalik District of southeast Greenland. Medd. Gr0enl. 198(6): 1-30.
NOVRUZOVA. Z. A. 1968. The water conducting system of trees and shrubs in relation to ecology. Iztatel'tsvo Akademii Nauk Azerbaijan S. S. R.. Baku.
SCHOLANDER. P. F . . H . H A M M I I , E. D . BRADSTREET, AND
E. A. HEMMINGSEN. 1965. Sap pressure in vascular plants. Science 148: 338-345.
STERN. W. L.. AND S. GREENE. 1958. Some aspects of
variation in woods. Trop. Woods 108: 65-71. VFRSTEEGH, C. 1968. An anatomical study of some
woody plants of the mountain flora in the tropics (Indonesia). Acta Bot. Neerl. 17: 151-159.
WEBBER, I. E. 1936. The woods of sclerophyllous and desert shrubs and desert plains oi" California. Amer. J. Bot. 23: 181-188.