“Secondary Stem Lengthening” in the Palm Iriartea Deltoidea (Arecaceae) Provides an Efﬁcient and Novel Method for Height Growth in a Tree Form

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/221878251

“Secondary stem lengthening” in the palm Iriartea deltoidea (Arecaceae) provides an efﬁcient and novel method for height growth in a tree form

Article in American Journal of Botany · March 2012 DOI: 10.3732/ajb.1100523 · Source: PubMed

CITATIONS READS 8 66

2 authors:

Heidi Renninger Nathan Phillips Mississippi State University Boston University

37 PUBLICATIONS 432 CITATIONS 88 PUBLICATIONS 7,118 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Research at the Wind River Canopy Crane Research Site View project

All content following this page was uploaded by Heidi Renninger on 05 June 2014.

The user has requested enhancement of the downloaded file. American Journal of Botany 99(4): 607–613. 2012.

“ S ECONDARY STEM LENGTHENING ” IN THE PALM I RIARTEA DELTOIDEA (ARECACEAE) PROVIDES AN EFFICIENT AND NOVEL METHOD FOR HEIGHT GROWTH IN A TREE FORM 1

H EIDI J. RENNINGER 2 AND N ATHAN P HILLIPS

Department of Geography and Environment, Boston University, 675 Commonwealth Avenue, Boston, Massachusetts 02215 USA

• Premise of the study: Although traditionally assumed that all height growth in trees occurs at apical meristems, sequential measurement of internode lengths in the palm Iriartea deltoidea suggested that stems were lengthening long after the differentiation of tissues and far below the apical meristem. This observation is diffi cult to reconcile with the fact that neither the water- conducting vessels nor the sugar-transporting sieve tube cells are capable of lengthening after differentiation. However, the vascular bundles in palms form a spiral within the stem and could theoretically lengthen if the spiral “ straightened ” . • Methods: We marked stretches of internodes on small and medium-sized palms and measured their lengths over 2 years. Ad- ditionally, we collected material from small palms with short internodes and large palms with long internodes and made cross sections to determine the angle of vascular bundles within stems. • Key results: We found that stems lengthened (up to 12% over 2 years) below the apical meristem in small and medium-sized palms and that the spiral angle in vascular bundles of small palms was signifi cantly larger than at the base of large palms indi- cating a straightening of the spiral. • Conclusions: These results represent the fi rst determination of “ secondary lengthening ” in tree stems as well as the most effi - cient method for height growth in terms of carbon investment. Likewise, elongation of stems allows palms to exhibit plasticity in height growth rates for more rapid growth when short-lived canopy gaps are present than they would have with apical growth alone.

Key words: Amazon rainforest; Arecaceae; growth rates; Iriartea deltoidea ; palms; tree height; water transport.

Despite the advantages of the tree growth form and its inde- single apical meristem from which all new leaves and trunk pendent evolution in a number of plant groups (Petit and Hampe, growth arise. One of the main consequences of the tree growth 2006 ), only one mechanism has been recognized for the growth habit is the necessity for long-distance water and solute trans- in height of trees, namely, cellular division and expansion at port (Sperry, 2003) and the sustained continuity of these sys- apical meristems. Trees compete in a race to the sky to gain the tems throughout the lifetime of the tree. It has been traditionally upper hand in light interception ( Iwasa et al., 1985 ; King, 1990 ; assumed that once the vascular conduits in stems are fully dif- Falster and Westoby, 2003). Although not the only strategy in ferentiated, no new elongation can occur because these cells are forested ecosystems, being tall is particularly important in dense highly lignifi ed and undergo either programmed cell death (in environments such as tropical rainforests (Clark and Clark, the case of xylem conduits) or lose their nucleus and most cel- 2001 ; Falster and Westoby, 2003 ) where light levels drop sig- lular organelles (in the case of phloem sieve tube cells) when nifi cantly from the top of the canopy to the ground (Canham fully differentiated (Greenberg, 1996). This process greatly re- et al., 1990 ; Kumagai et al., 2001 ; Montgomery and Chazdon, duces maintenance costs of the tree (Sperry, 2003), but the 2001). By achieving signifi cant heights, trees also avoid ground conduits have little ability to signifi cantly elongate after differ- herbivores and falling debris (Sterck and Bongers, 1998), and entiation is completed. their wind-dispersed pollen and seeds can be transported farther Unlike diameter growth, which is easily tracked in temperate ( Ruokolainen and Vormisto, 2000 ; Petit and Hampe, 2006 ). dicotyledonous trees, height growth is much more diffi cult to Trees grow taller through cell division and expansion at api- determine across the lifetime of an individual. Likewise, many cal meristems located at the ends of the main shoot and branches. tropical trees lack distinct annual growth rings, which limits Palm trees grow taller in a similar fashion, but they maintain a the ability to attach a chronological scale to past events (Mart í nez-Ramos et al., 1988). However, in many species of palm trees, it is relatively easy to track height growth because 1 Manuscript received 1 November 2011 revision accepted 25 January 2012. the nodal scars that are left when a leaf is produced and later The authors thank the staff at Universidad San Francisco de Quito, lost are clearly visible along the trunk. If the length of time Tiputini Biodiversity Station (TBS), D. Mosquera for data collection, and between the production of successive leaves is also known, Dr. T. Blute for technical advice and microscopy assistance. Funding height growth rates can be estimated ( Tomlinson, 1963 ; Lugo was provided by a grant from the National Science Foundation (IOB and Batlle, 1987; Renninger and Phillips, 2010) and have been #0517521). 2 Author for correspondence (e-mail: [email protected]); present used, for instance, to age treefall gaps in a tropical forest address: 195 University Ave., Rutgers University, Newark, New Jersey (Mart í nez-Ramos et al., 1988). The distance between nodes on 0710 USA; 857-488-5144 the trunk of palms is assumed to be fi xed once produced (Waterhouse and Quinn, 1978; Lugo and Batlle, 1987). How- doi:10.3732/ajb.1100523 ever, when measuring internode distances in Iriartea deltoidea

(Ruiz & Pav.; Arecaceae) palms in Amazonian Ecuador, we MATERIALS AND METHODS found evidence which suggested that internodes elongated far below the base of the live crown and long after vascular con- Measurement of internode sizes — Iriartea deltoidea palms were sampled duits were fully differentiated (Renninger and Phillips, 2010). growing within Tiputini Biodiversity Station (0° 36 ′ S, 76 ° 27 ′ W), a 650-ha pri- Internodes in short palms were always signifi cantly shorter than mary rainforest facility located within Yasuní National Forest in eastern Ecua- internodes at the base of tall palms. Likewise, internodes at the dor. In March 2008, aluminum tags were hammered into the most recently formed internode in four palms (three 1-m tall palms and one 12-m tall palm). tops of medium-sized palms (still well below maximum heights All internodes lengths were measured and summed from this tag to the lowest seen in this species) were much shorter than at midheight on tall identifi able internode. In November 2010, internode lengths from these tags to palms. These patterns could be the result of differences in grow- the lowest identifi able internode were remeasured and summed, making sure ing environment between different individuals during the same that the number of internodes was the same as 2008. Additionally, in two palms life stages, but the climatic stability of this region of the Ecua- (4 m and 9 m tall), internodes were measured in March 2008 from the lowest dorian rain forest argues against this. identifi able internode to the highest internode that could be reached using a 3-m tall ladder. These palms were revisited in November 2010, and their internodes Although data suggest that I. deltoidea boles can elongate were remeasured. Matching up the patterns between individual internodes and well below the crown after differentiation, the vascular bundles the internode counts confi rmed that the same location was measured in both need a mechanism to maintain continuity after elongation. instances. Figure 2 presents photographs of representative internodes from While it is unlikely that the individual vascular conduits can small palms and from the base and top of medium and tall palms. lengthen signifi cantly, the vascular bundles in palms and the conduits within them form a spiral as they travel up the bole Microscopy — Samples from the boles of I. deltoidea palms were collected ( Zimmermann and Tomlinson, 1965 ; Zimmermann, 1972 ). Thus, in January and February 2007 from four small palms with trunks between 0.5 m the vascular bundles could conceivably lengthen after differen- and 3 m long, one medium-sized palm with a 14 m long trunk and one tall palm with a 20 m long trunk. The small palms were harvested to access the samples, tiation if this spiral elongates and narrows ( Fig. 1 ). On refl ec- while the 14 m and 20 m tall palms had been pushed over in treefalls allowing tion, it is evident that palms — indeed all trees — must possess samples from midbole and the top to be accessed. Palms differed from those vertical vascular elasticity, to the extent that they bend revers- used to monitor internode elongation. Four subsamples (two from the inner and ibly in the wind (Winter, 1993). Therefore, we seek to demon- two from the outer portion of the central cylinder) per palm were taken at mid- strate that internodes in I. deltoidea lengthen after conduits height from the trunks of the small palms. For the 14 m and 20 m tall palms, have fully differentiated and show that “ straightening ” of the samples were taken from around midheight and from just below the base of the live crown. Again four subsamples were taken from each height location — two vascular bundles facilitates this lengthening. If short inter- from the inner and two from the outer portion of the central cylinder. nodes have vascular bundles with a greater spiral angle (from From each sample, cross sections were hand-cut using a razorblade, stained vertical) than large internodes, this will provide evidence in a solution of 1% toluidine blue O, destained in water, and placed on a cover- that I. deltoidea stems have vascular systems with the capacity slip. Any excess water was allowed to dry, and Permount (Fisher Scientifi c, for internode elongation after differentiation (i.e., “ second- Fair Lawn, New Jersey, USA) was applied. Another coverslip was placed over ary” lengthening), which represents a heretofore unrecognized the fi rst making a “ sandwich ” around the palm bole sections. After the Per- mount had dried completely, a fi ne-tip permanent marker was used to make method of achieving height growth in a tree form. dashed lines around the edges of the sections that would act as reference points in the photographs. The sections were viewed using an Olympus IX70 inverted microscope (Center Valley, Pennsylvania, USA) with a 4× objective and an Optronics MagnaFire CCD camera (Goleta, California, USA) attached. A photograph was taken of the “ top ” of the section, then the coverslips were fl ipped, and another photograph was taken of the “ bottom ” of the same section. This was repeated to generate photograph pairs of several different locations for each subsample. The photographs were printed onto transparency sheets, and the photograph pairs (“ top ” and “ bottom ” ) were matched up using the edges of the sections and the marker lines ( Fig. 3 ). The matched photographs were placed onto an over- head projector and magnifi ed onto a screen. The amount of offset ( “x ” in Fig. 1 ) between the “ top ” and “ bottom ” of an individual vessel was measured on all clearly visible vessels in each photograph pair (Fig. 3). With this “ offset ” distance and the assumptions that the hand sections were ca. 25 µ m thick and were cut perpendicular to the vertical growth plane, trigonometry was used to determine the angle of the vessels ( “θ ” in Fig. 1 ) within the samples (with 0° repre- senting vertical).

Statistical analyses — Measured vessel angles within each of the four subsamples from each location were averaged. Then, averages from each subsample were compared using an ANOVA where vessel angle was the dependent variable and sample location (palm size/bole location) was the predictor variable. Multiple comparisons between all measured locations were made using Tukey honestly signiﬁ cant difference (HSD) tests in the program R version 2.5.1 (R Foundation for Statistical Computing, http://www.R-project.org).

RESULTS Fig. 1. Schematic of the proposed lengthening of vascular bundles in Iriartea deltoidea palms. Bundles get longer ( y ) through a decrease in the Over 32 months, Iriartea deltoidea palm stems lengthened angle (θ ) of the spiral (from vertical) and measured empirically by the ca. 12% (SE = 0.036, N = 3) in small palms and ca. 3% (SE = amount of offset (x ) in an individual vessel at two vertical locations within 0.0044, N = 3) in medium-sized palms as a result of internode a cross section (i.e., the “ top ” and “ bottom ” of the section). lengthening alone (Fig. 4). Vessel angles from the palm height/ April 2012] RENNINGER AND PHILLIPS — SECONDARY LENGTHENING IN PALMS 609

Fig. 2. Internode distances in Iriartea deltoidea individuals from (A) a small palm (ca. 1 m tall), the top of (B) a medium-sized palm (ca. 10 m tall) and (C) a tall palm (ca. 22 m tall) and the lower bole of (D) a medium-sized palm (ca. 12 m tall), and (E) a tall palm (ca. 22 m tall). Lines in each photograph give approximate lengths of a typical internode from each palm size/location. 610 AMERICAN JOURNAL OF BOTANY [Vol. 99

Fig. 3. Superimposed photograph pairs of the “ top ” and “ bottom ” of cross sections from the boles of Iriartea deltoidea palms used to identify vessel angles in samples taken from (A) midheight in a short palm, (B) midheight in a 14 m tall palm, (C) just below the crown of a 14 m tall palm, (D) midheight on a 20 m tall palm, and (E) just below the crown of a 20 m tall palm. White arrows are located on the marker lines that are matched in the photograph pairs, and black arrows point to the measured “ offset ” in vessels corresponding to their angle from vertical. The scale bar represents 500 µ m.

location categories differed signiﬁ cantly (F 4, 21 = 23.5, P < than locations with longer internodes (Fig. 5). Vessel angles in 0.001) from one another. Except for the top of the tallest palm, small palms with short internodes were ca. 41.7° (SE = 0.4) from bole locations with shorter internodes had signiﬁ cantly larger vertical, while vessel angles at midheight in a 14 m and 20 m tall vessel angles, and therefore a tighter vascular bundle spiral, palm (where internodes are much larger), were 20.1° (SE = 0.7) April 2012] RENNINGER AND PHILLIPS — SECONDARY LENGTHENING IN PALMS 611

Fig. 5. Calculated vessel angles (from vertical) for the various Iriartea deltoidea palm height – bole location combinations. Black bars are averages of samples from midheight; gray bars are averages of samples from just below the crown. Standard errors represent the variation in four subsamples for each location in medium and tall palms and four subsamples from each of four small palms. Bars with different letters differ signiﬁ cantly from one another at P = 0.01 based on post hoc Tukey HSD tests.

Fig. 4. Vertical distance between the same stretch of “n ” internodes can occur well below the live crown in a palm species by ca. ( “ n” in white) measured in March 2008 (black bars) and November 2010 12% in small palms and 3% in medium-sized palms over (black and gray bars) in six Iriartea deltoidea palms. 32 months. These numbers are signifi cant, considering this lengthening occurred over a relatively short period compared with the overall lifetime of these palms. Likewise, in the medium-sized and 16.8 ° (SE = 2.3) from vertical, respectively (Fig. 5). Post palms, this amount of lengthening is comparable to the height hoc Tukey HSD tests showed that vessel angles in small palms growth that would occur in this time period due to apical growth differed signifi cantly from midheight on the 14 m tall palm alone assuming one or two leaves were produced annually ( P < 0.001) and midheight on the 20 m tall palm ( P < 0.001). (Renninger and Phillips, 2010). In order for internodes to lengthen, Likewise, large vessel angles of 33.8° (SE = 0.8) were seen at the vascular bundles and the dead cells within them need to the top of the 14 m tall palm where small internodes were also lengthen to maintain continuity. We have found that the angle present ( Fig. 5 ). Again, post hoc Tukey HSD tests showed that of the vascular bundles (from vertical) in short internodes is vessel angles at the top of the 14 m tall palm were signifi cantly much greater than in larger internodes, suggesting that they larger than at midheight on this same palm (P = 0.008) signifi - lengthen in a similar way that a spring lengthens; by the elonga- cantly larger than midheight on the 20 m tall palm (P < 0.001) tion and narrowing of their spiral pattern. Our analysis of vas- but not signifi cantly different from vessel angles in the small cular bundle angles shows that internodes can potentially triple palms ( P = 0.68). However, at the top of the 20 m tall palm with in length by decreasing spiral angles (given that large palms similarly small internodes, vessel angles were about 16.1° (SE have angles that are ca. 1/3 the degree of angles seen in small = 2.0) and did not differ signifi cantly from either midheight on palms; Fig. 5). This degree of length change in internodes is on that same palm (P = 0.99) or from midheight on the 14 m tall the same order of magnitude as differences in internode sizes palm ( P = 0.80) both of which have much larger internodes. seen between small palms and the bases of large I. deltoidea However, vessel angles at the top of the 20 m tall palm were palms (Renninger and Phillips, 2010). It is also interesting to signifi cantly smaller than the top of the 14 m tall palm (P < note that the short internodes at the top of the 20 m tall palm had 0.001) and from the small palms (P < 0.001) based on post hoc small vascular bundle angles and, therefore, little ability to Tukey HSD tests. elongate further under this mechanism. However, this palm is nearing the maximum height for individuals of this species ( Henderson, 1990 ) and is likely shutting down its height growth DISCUSSION rates with no need for further stem elongation. Iriartea deltoidea is one of the most abundant tree species in Although it has been traditionally assumed that internodes in the northwestern Amazon (Pitman et al., 2001; Montufar and palms cannot elongate after leaf fall ( Branner, 1884 ; Waterhouse Pintaud, 2006 ), and its capacity for secondary stem elongation may and Quinn, 1978 ), we have shown that lengthening of the bole contribute to this. Typically, palms get taller by adding new 612 AMERICAN JOURNAL OF BOTANY [Vol. 99 leaves and stem tissue at their apical meristem. Therefore, for a sustained primary growth and are hypothesized to have second- species like I. deltoidea , which holds less than 10 leaves per palm ary elongation of their stems as well. Secondary lengthening as and replaces between one to four leaves per year ( Renninger a mode of height growth is highly effi cient because it requires and Phillips, 2010 ), height growth rates would be relatively no additional carbon to be invested ( King, 1981 ), giving species slow through apical growth alone. Indeed, apical growth in with this ability an advantage over species with only primary I. deltoidea could not outcompete many tropical angiosperm apical growth. This fi nding also highlights the fact that in addi- species considering they are capable of growing 1– 3 m in height tion to biodiversity, tropical rainforests also contain a high de- per year (King, 1993; Clark and Clark, 2001). Of course, rapid gree of “ physiological ” diversity, the depth of which is just height growth is not the only successful strategy in rainforests, beginning to be discovered. and many species are adapted to spending their entire lives in the understory. However, studies have found that the most common and widespread palm species throughout the Amazonian LITERATURE CITED region tend to be those that are tall and reach the canopy A VALOS , G . , AND M . F ERN Á NDEZ OT Á ROLA . 2010 . Allometry and stilt (Ruokolainen and Vormisto, 2000; Kristiansen et al., 2009). root structure of the neotropical palm Euterpe precatoria (Arecaceae) Likewise, because the production of new leaves requires an in- across sites and successional stages. American Journal of Botany 97 : crease in stem height, palms are required to continuously get 388 – 394 . taller throughout their lifetimes ( Avalos and Fern á ndez Ot á rola, B RANNER , J. C. 1884 . The course and growth of the fi bro-vascular bundles 2010 ). In fact, many palms reach signifi cant heights even when in palms. Proceedings of the American Philosophical Society 21 : there are no other tall individuals competing with them for light. 459 – 483 . The phenomenon of “ secondary lengthening ” that we ob- C ANHAM , C. D. , J. S. DENSLOW , W. J. PLATT , J. R. RUNKLE , T. A. SPIES , AND P. S. WHITE . 1990 . Light regimes beneath closed canopies and tree- served in I. deltoidea , may also apply to other common palm fall gaps in temperate and tropical forests. Canadian Journal of Forest species as well. I. deltoidea exhibits diffuse secondary growth Research 20 : 620 – 631 . ( Tomlinson, 1961 ) or sustained primary growth ( Waterhouse C LARK , D. A. , AND D. B. CLARK . 2001 . Getting to the canopy: Tree height and Quinn, 1978 ; Tomlinson et al., 2011 ) where division and growth in a neotropical rain forest. Ecology 82 : 1460 – 1472 . expansion of the parenchyma cells and lacunae (Rich, 1987) F ALSTER , D. S. , AND M . W ESTOBY . 2003 . Plant height and evolutionary within the bole (as seen in Fig. 3B, D compared to Fig. 3C, E ) games. Trends in Ecology & Evolution 18 : 337 – 343 . result in signifi cant increases in diameter once established (up G REENBERG , J. T. 1996 . Programmed cell death: A way of life for plants. to 5-fold in our study trees). Palms with this growth type, in- Proceedings of the National Academy of Sciences, USA 93 : 12094 – 12097 . cluding Euterpe precatoria Mart., Euterpe edulis Mart., Pres- H ENDERSON , A. 1990 . Arecaceae part I. Introduction and the Iriarteinae. toea decurrens (H.Wendl. ex Burret) H.E.Moore, Prestoea New York Botanical Garden, Bronx, New York, USA. I WASA , Y . , D . C OHEN , AND J. A. LE Ó N . 1985 . Tree height and crown shape, acuminata (Willd.) H.E.Moore, Socratea exorrhiza (Mart.) as results of competitive games. Journal of Theoretical Biology 112 : H.Wendl ( Rich, 1986 ; Avalos and Fern á ndez Ot á rola, 2010 ), 279 – 297 . and Archontophoenix cunninghamiana H.Wendl. & Drude K ING , D. 1981 . Tree dimensions: Maximizing the rate of height growth in (Waterhouse and Quinn, 1978), therefore already exhibit a large dense stands. Oecologia 51 : 351 – 356 . degree of vascular rearrangement in terms of diameter growth K ING , D. A. 1990 . The adaptive signifi cance of tree height. American (see Tomlinson et al., 2011. fi gs. 16, 17), and it seems feasible Naturalist 135 : 809 – 828 . that their internodes would elongate as well. The other main K ING , D. A . 1993 . Growth history of a Neotropical tree inferred from the growth pattern in palms involves underground increases in di- spacing of leaf scars. Journal of Tropical Ecology 9 : 525 – 532 . ameter, with an aboveground trunk emerging only after a fi nal K RISTIANSEN , T . , J . C . S VENNING , C . G R Á NDEZ , J . S ALO , AND H. BALSLEV . 2009 . Commonness of Amazonian palm (Arecaceae) species: Cross- diameter is achieved. Measuring internodes in a palm species with scale links and potential determinants. Acta Oecologica 35 : 554 – 562 . this growth pattern (Mauritia fl exuosa L.f.) suggested that its inter- K UMAGAI , T . , K . K URAJI , H . N OGUCHI , Y . T ANAKA , K . T ANAKA , AND M . nodes do not elongate below the live crown ( Renninger and S UZUKI . 2001 . Vertical profi les of environmental factors within Phillips, 2010 ) and that the phenomenon of stem lengthening tropical rainforest, Lambir Hills National Park, Sarawak, Malaysia. may be limited to palms exhibiting sustained primary growth. Journal of Forest Research 6 : 257 – 264 . We have shown that I. deltoidea palms have the ability to L UGO , A. E. , AND C. T. R. BATLLE . 1987 . Leaf production, growth rate, lengthen their stems far below the apical meristem and maintain and age of the palm Prestoea montana in the Luquillo Experimental vascular connectivity by elongating the spiral of their vascular Forest, Puerto Rico. Journal of Tropical Ecology 3 : 151 – 161 . bundles. To our knowledge, this fi nding is the fi rst evidence of M ART Í NEZ-RAMOS , M. , E . A LVAREZ-BUYLLA , J . S ARUKH Á N , AND D . P I Ñ ERO . 1988 . Treefall age determination and gap dynamics in a tropical for- a tree that exhibits “ secondary lengthening” of its stem. The est. Journal of Ecology 76 : 700 – 716 . spiral of the vascular bundles in stems prior to elongation is M ONTGOMERY , R. A. , AND R . L . C HAZDON . 2001 . Forest structure, canopy much tighter with vascular bundles angles much more inclined architecture, and light transmittance in tropical wet forests. Ecology from vertical, allowing stems to elongate via a narrowing of this 82 : 2707 – 2718 . spiral without compromising the connectivity of the system. In- M ONTUFAR , R . , AND J. C. PINTAUD . 2006 . Variation in species composition, deed, having large angles in the vascular bundle spiral adds abundance and microhabitat preferences among western Amazonian path length resistance to fl uid transport that would make this an terra fi rme palm communities. Botanical Journal of the Linnean ineffi cient system hydraulically if additional stem lengthening Society 151 : 127 – 140 . was not the ultimate goal. Therefore, the ability for secondary P ETIT, R. J., AND A . H AMPE. 2006 . Some evolutionary consequences of being a stem elongation appears to be important for I. deltoidea growing in tree. Annual Review of Ecology Evolution and Systematics 37 : 187 – 214 . P ITMAN , N. C. A. , J. W. TERBORGH , M. R. SILMAN , P . NÚ Ñ EZ , D. A. NEILL , rainforest environments where plasticity in height growth rates C. E. C ER Ó N, W. A. P ALACIOS , ET AL. 2001 . Dominance and distribution and rapid height growth are important for taking advantage of of tree species in upper Amazonian terra fi rme forests. Ecology 82 : short-lived canopy gaps. Indeed, some of the most common palm 2101 – 2117 . species in western Amazonia (I. deltoidea , Socratea exorrhiza , R ENNINGER, H. J., AND N . P HILLIPS. 2010 . Intrinsic and extrinsic hydraulic and Euterpe precatoria) (Kristiansen et al., 2009) all have factors in varying sizes of two Amazonian palm species (Iriartea deltoidea April 2012] RENNINGER AND PHILLIPS — SECONDARY LENGTHENING IN PALMS 613

and Mauritia ﬂ exuosa) differing in development and growing environ- T OMLINSON , P. B. 1961 . Anatomy of the monocotyledons. II. Palmae. ment. American Journal of Botany 97 : 1926 – 1936 . Clarendon Press, Oxford, UK. R ICH , P. M. 1986 . Mechanical architecture of arborescent rain forest T OMLINSON , P. B . 1963 . Measuring growth rates in palms. Principes 7 : palms. Principes 30 : 117 – 131 . 40 – 44 . R ICH, P. M. 1987 . Developmental anatomy of the stem of Welfia T OMLINSON , P. B. , J. W. HORN , AND J. B. FISHER . 2011 . The anatomy of georgii, Iriartea gigantea and other arborescent palms: Implica- palms. Oxford University Press, Oxford, UK. tions for mechanical support. American Journal of Botany 7 4 : W ATERHOUSE , J. T. , AND F. L. S. QUINN . 1978 . Growth patterns in the stem 792 – 802 . of the palm Archontophoenix cunninghamiana. Botanical Journal of R UOKOLAINEN , K. , AND J . V ORMISTO . 2000 . The most widespread the Linnean Society 77 : 73 – 93 . Amazonian palms tend to be tall and habitat generalists. Basic and W INTER, D. F. 1993 . On the stem curve of a tall palm in a strong wind. Applied Ecology 1 : 97 – 108 . SIAM Review 35 : 567 – 579 . S PERRY , J. S. 2003 . Evolution of water transport and xylem structure. Z IMMERMANN , M. H. 1972 . The vascular system of monocotyledonous International Journal of Plant Sciences 164 : S115 – S127 . stems. Botanical Gazette 133 : 141 – 155 . S TERCK , F. J. , AND F . BONGERS . 1998 . Ontogenetic changes in size, al- Z IMMERMANN , M. H. , AND P . B . T OMLINSON . 1965 . Anatomy of the lometry, and mechanical design of tropical rain forest trees. American palm Rhapis excelsa I. Mature vegetative axis. Journal of the Arnold Journal of Botany 85 : 266 – 272 . Arboretum 46 : 160 – 181 .

View publication stats