Bradleya 24/2006 pages 39 – 52
Pollination biology of three Brazilian species of Micranthocereus Backeb. (Cereeae, Cactoideae) endemic to the “campos rupestres” Lidyanne Yuriko Saleme Aona 1, Marlon Machado 2 , Emerson Ricardo Pansarin 1, Cibele Cardoso de Castro 3 , Daniela Zappi 4 and Maria do Carmo E. do Amaral 1 Dept. de Botânica, IB, Universidade Estadual de Campinas, Caixa Postal 6109, Cidade Universitária s.n., 13084-970, Campinas-SP, Brazil (email:
[email protected]). 2 University of Zurich, Institute for Systematic Botany, Zollikerstrasse 107, CH-8008, Zurich, Switzerland. 3 Departamento Departament o de Biologia, Universidade Federal Rural de Pernambuco R. Dom Manoel de Medeiros, s/n - Dois Irmãos, 52171-900, Recife-Pernambuco, Brazil. 4 Royal Botanic Gardens, Kew, Richmond, Surrey, TW9 3AE, UK. 1
Summary: The
genus Micranthocereus includes nine endemic Brazilian species distributed in the t he states of Bahia, Minas Gerais, Goiás and Tocantins. In the present study the floral biology of M. flaviflorus , M. steckeri and M. purpureus from “campo rupestre” vegetation of the Chapada Diamantina, Bahia, was investigated. Data on floral morphology and nectar production (concentration and volume) were collected throughout anthesis. Flowers of all species are tubular and present coloured perianth-segments. Nectar is accumulated within a nectar-chamber, located inside the floral tube. Flowers of M. flaviflorus and M. streckeri are predominantly nocturnal, opening in the night (at about 7 p.m.). Each flower lasts c.36 hours, with a concentration of sugar in the nectar between approximately 18–24%. Both species present floral features typical for ornithophily, and were pollinated by the hummingbird Clorostilbon aureoventris , whose visits occurred in regular intervals and were characterized by territorialist t erritorialist behaviour. behaviour. Species of bees and butterflies were also observed visiting flowers of M. flaviflorus and M. streckeri . M. streckeri was additionally visited by the hummingbird Phaethornis squalidus , mainly in the afternoon. Flowers of M. purpureus are nocturnal, and each flower lasts c.17 hours and present longer corollas with wider diameter than those of the other two species studied; sugar concentration in the nectar was about 21–22%. Three
sphingid and one bat species were observed visiting flowers of M. purpureus , with a greater frequency of visits of the sphingids. Pollen viability was high (80–90%) and similar between the three species of Micranthocereus studied. Zusammenfassung : Die thocereus umfasst neun in
Gattung MicranBrasilien endemische Arten, und ist in den Bundesstaaten Bahia, Minas Gerais, Goiás und Tocantins Tocantins verbreitet. In der vorliegenden Arbeit wurde die Blütenbiologie der Arten M. fl flav avif iflo loru rus s , M. purpureus und M. streckeri aus den “campos rupestres” (Felssavannen) der Chapada Diamantina (Bahia) untersucht. Während der gesamten Blühphase wurden Daten über Blütenmorphologie und Nektarproduktion (Konzentration und Volumen) Volumen) gesammelt. Die Blüten aller Arten sind röhrenförmig and haben ein farbiges Perianth. Der Nektar sammelt sich in der Nektarkammer, die sich im Inneren der Blütenröhre befindet. Die Blüten von M. flaviflorus and M. streckeri sind hauptsächlich nachts geöffnet. Die Anthese beginnt abends gegen ungefähr 19.00. Eine Blüte dauert etwa 36 Stunden und hat eine Zuckerkonzentration im Nektar zwischen 18–24%. Beide Arten weisen Blütenmerkmale auf, die typisch für Vogelblumen sind. Sie wurden von der Kolibriart Clorostilbon aureoventris bestäubt, dessen Besuche in regelmässigen Intervallen erfolgten und durch
territoriales Verhalten gekennzeichnet waren. Ausserde Auss erdem m wurde wurden n vers verschie chiedene dene Arte Arten n von Bienen und Schmetterlingen bei Blütenbesuchen von M. flaviflorus und M. streckeri beobachtet. Besonders am Nachmittag wurde M. streckeri zusätzlich von der Kolibriart Phaethornis squalidus besucht. M. purpureus ist nachtblühend, und jede Blüte bleibt etwa 17 Stunden geöffnet. Ihre Blütenröhren sind länger und weisen auch einen grösseren Durchmesser auf als diejenigen der beiden anderen Arten. Die Zuckerkonzentration des Nektars betrug 21–22%. Drei Arten von Schwärmern und eine Fledermausart wurden bei Blütenbesuchen beobachtet, wobei die Schwärmer häufigere Besuche zeigten als die Fledermäuse. Die Lebensfähigkeit des Pollens aller drei untersuchten Arten war hoch und betrug ca. 80–90%. Resumo :
O gênero Micranthocereus inclui nove espécies endêmicas do Brasil e restritas aos Estados da Bahia, Minas Gerais, Goiás e Tocantins. Foi estudada a biologia floral de M. flaviflorus , M. steckeri e M. purpureus em campos rupestres na Chapada Diamantina, BA. Análises da morfologia floral e medições do volume e concentração do néctar desde a antese até o pré-fenescimento das flores foram realizadas. As flores das três espécies estudadas são tubulares, apresentando segmentos do perianto diversamente coloridos. O néctar é armazenado na câmara nectarífera, localizada no interior do tubo floral. As flores de M. flaviflorus e M. streckeri apresentam antese noturna, abrindo no início da noite (aproximadamente 7 horas da noite). Cada flor dura cerca de 36 horas e apresenta concentração de açúcares entre 18 e 24%. Ambas as espécies apresentam características típicas de flores polinizadas por aves e são visitadas e polinizadas pela mesma espécie de beija-flor beija-flor,, Clorostilbon aureoventris . As visitas ocorreram em intervalos regulares e o beija-flor apresentou comportamento territorialista, quando dois indivíduos visitavam simultaneamente as flores. Visitas de abelhas e borboletas também foram observadas nestas espécies. M. streckeri também foi visitado por outro beija-flor, Phaethornis squalidus , que efetuou um maior número de visitas no período vespertino. As flores de M. purpureus apresentam antese noturna, e cada flor dura aproximadamente 17 horas. O comprimento e diâmetro floral são superiores aos das espécies
diurnas e a concentração de açúcares varia entre 21 e 22%. Foram observadas visitas de três espécies de esfingídeos e de uma espécie de morcego. Foi observada uma maior freqüência de visitas dos esfingídeos em relação ao morcego. O teste de viabilidade polínica foi semelhante entre as espécies estudadas (80–90%). Introduction The family Cactaceae forms a monophyletic group (Wallace & Cota, 1996) with approximately 100 genera and 1300 species distributed nearly exclusively in the New World (Barthlott & Hunt, 1993). Eastern Brazil presents the third main geographical centre of diversity of the family (Zappi, 1994; Taylor & Zappi, 2004). The core geographical area defined within Eastern Brazil has been recognized primarily for its remarkable endemism of Cactaceae genera and species, especially those of the “caatingas” of the North-eastern states and associated highland known as “campos rupestres” (Taylor & Zappi, 2004). Cactaceae present a wide range of floral features regarding shape, colour, size, odour, time of anthesis and position of the reproductive parts, which indicate pollination by several groups of biotic agents (Porsch, 1939; 1939 ; Hunt & Tayl Taylor or,, 1990; Vogel, V ogel, 1990), like medium-sized (McFarland et al. , 1989) and large bees (Viana et al., 2001), sphingids and other moths (Suzan et al., 1994; Holland & Fleming, 1999), hummingbirds (Scobell & Scott, 2002), bats (Ruiz et al., 1997; Tschapka et al., 1999), or even nocturnal and diurnal vectors in the same species, such as by hummingbirds and bats (Fleming et al., 1996; Sahley, 2001). According to Vogel (1990), around 9% of the species of Cactaceae are pollinated by hummingbirds. Other studies on Cactaceae also reported several reproductive strategies like selfcompatibility (Bianchi et al., 2000) and selfincompatibility (Boyle, 1996; Metz et al., 2000), as well as the presence of sexual and asexual reproduction in the same species (Mandujano et ., 1998; Negron-Ortiz, 1998). al ., The Brazilian genus Micranthocereus Backeb. (Cactoideae, Cereeae) includes nine species distributed in the states of Minas Gerais, Bahia, Goiás and Tocantins (Taylor & Zappi, 2004). Studies concerning the reproductive biology of Brazilian Cactaceae recorded melitophily in Opuntia Mill. (Schlindwein & Wittmann, 1995, 1997), sphingophily in Cereus Mill. (Silva & Sazima, 1995; Locatelli & Machado, 1999a) and Selenicereus (Berg.) Britton & Rose (Barthlott et
al. ,
1997), orn 1997), ornith ithoph ophily ily in Melocactus Link & Otto and Opuntia (Locatelli & Machado, 1999b; Raw, 1996), and chiropterophily in Pilosocereus Byles & Rowley (Locatelli et al .,., 1997). The reproductive biology of Micranthocereus species is unknown, but the genus presents great diversity in flower morphology. The main goal of this study was to investigate and compare the pollination biology of three Brazilian Micranthocereus species: M. flaviflorus Buining & Brederoo, M. streckeri van Heek & van Criekinge and M. purpur ure eus (Gürke) Ritter endemic to northeast Brazil. Material and Methods The floral biology of the species studied was investigated in the Chapada Diamantina (Bahia State), a mountain complex that presents a large diversity of Cactaceae species (Machado, 1999; Taylor & Zappi, 2004) and which is located in the Cadeia do Espinhaço, a mountain range which extends over c.1000 km from NE to SE Brazil. The study was conducted in August 2001 for M. flaviflorus and M. streckeri and in October 2003 for M. purpureus . The studied population of M. flaviflorus consisted of c. 20 individuals located in the municipal district of Morro do Chapéu (10°50’S–12°00’S 1 and 40°40’W–41°30’W) . The plants grow in quartzitic sand in an area occupied by transitional vegetati vegetation on includ including ing ‘cerra ‘cerrado’ do’ (sava (savanna) nna),, ‘campoo rupestre’ ‘camp rupestre’ (upl (upland and or highland highland savanna) savanna) and ‘caatinga’ ‘caatinga’ (deci (deciduou duouss low thorn forest). The population of M. streckeri consisted of 40 individuals and was located upon rocks in an area of ‘campo rupestre’ rupestre’ vegetation surrounded surrounded by forest areas, being endemic to the study area, in the Municipality of Seabra (12º25’52S and 41º59’36W). M. purpureus presents a much larger distribution range, occurring from Rio de Contas to Jacobina (Aona, 2003). The population studied consisted of 40 individuals and is located in the Municipality of Morro do Chapéu (11º36’03S and 41º09’52W), occurring in an area of ‘campo rupestre’ rupestre’ vegeta vegetation. tion. The growth habit, phenology and floral features, such as morphology morphology,, colour and scent, were recorded through field observation. Floral lifespan and anthesis events were recorded for a population of each species through the observation of 1
tagged flowers from pre-anthesis until flower senescence. Fresh flowers of each species were randomly collected in a minimum of 15–30 individuals/species, fixed in FA FAA A 50, stored in alcohol 70% and used for illustrations and for floral measurements (flower-tube length and diameter, 17–32 flowers/species). Floral measurements were obtained with a digital calliper (error 0.01 mm) and compared between species with an ANOVA ANOV A analysis coupled with a t-Test LSD (Sokal & Rohlf, 1969). Nectar concentration and volume were estimated in flowers bagged from the pre-anthesis stage, using a pocket refractometer and microliter syringes, respectively. respectively. Data were collected c. 24 hours after the beginning of the anthesis for M. flaviflorus (n=12 flowers) and M. streckeri (n=10 flowers, five individuals for both), and 3 hours (n=10, eight individuals) and 9 hours (n=8, six individuals) after anthesis for M. purpureus . Pollen viability was estimated using the acetatecarmine technique (Radford et al., 1974, 5-7 flowers/species, 200 pollen grains/flower). Floral visitors were recorded through direct observations in the field, in a total of 10h for M. flaviflorus , 40h for M. streckeri and 40h for M. purpureus . Time and duration of visiting bouts were recorded, as well as the visitor’s behaviour on the flowers, such as contact with the anthers and the stigma, type of floral resource collected and interactions with other visitors. Buds/mature flowers were tagged in the afternoon and examined early in the morning for detecting possible nocturnal pollination in M. flaviflorus and M. streckeri . The hummingbirds were identified in the field and using photographs taken during their visits, then compared with ilustrations in Ruschi (1986) and Grantsau (1989). Others floral visitors of M. purpureus were identified using photos taken during their visits to the flowers; these photos were also used to check the sites of pollen deposition on their bodies. Insect visitors of all species were collected and identified as far as possible through comparison with specimens deposited in the Museum of the Universidade Estadual of Campinas (ZUEC). Pollinators were characterized based on their frequency of visits and visiting behaviour. Frequency of visits of the main floral visitors of
This population belongs to the taxon recently recognized as M. polyanthus (Werderm.) Backeb. subsp. alvinii M. Machado & Hofacker. These plants are to some degree intermediate between M. flaviflorus and M. polyanthus (Werderm.) Backeb. which are Hofacker. similar species (Taylor & Zappi, 2004: 355).
3mm 3mm
Figure 1. 1. Lon Longit gitudi udinal nal secti section on of flo flower wer of of M. M. flavi- florus . Note the nectar nectar chamber chamber (arro (arrow). w). Illust Illustrat ration ion by R. R. Lup Lupo o.
Figure 2. Figure 2. Longit Longitudinal udinal sectio section n of flow flower er of of M. M. streck- eri . Note the nectar nectar chamber chamber (arro (arrow). w). Illust Illustrat ration ion by R. Lu Lupo po..
each species was compared using a Student’s t test (Sokal & Rohlf, 1969). Voucher V oucher specimens were deposited in the Herbarium of the Universidade Estadual de Campinas ( M. flaviflorus : UEC 120736, M. streckeri : UEC 139459 and M. purpureus UEC 132282).
(77–90%) which is similar between species (F=3.35, P=0.06). In M. flaviflorus and M. streckeri the fertile zone consists of a lateral, superficial to sunken cephalium with flower-bearing areoles, which have white to yellowish hairs with long, pale to brownish golden bristles in M. flaviflorus , and compact, white to brownish hairs and long, pale golden to reddish brown bristles in M. streckeri . In M. purpureus , the fertile portion of the stem is a lateral cephalium sunken into the stem and composed of congested flower-bearing areoles which bear pale brown, pinkish or grey, abundant, loose or compact hair and few long golden or brownish bristles. M. flaviflorus flowered from May to September and produced fruits from July to November, whereas M. streckeri produced flowers from May to September and fruits from August to December. Both species present an annual
Results Morphology and floral biology The plants of all studied species are columnar, branching at the base, with decumbent or erect stems and covered with flexible, golden to pale golden spines. Flowers are born in a specialized fertile zone (cephalium), more or less perpendicular and lateral to the stem axis. The flowers present a nectariferous chamber with a nectary inside the base of the flower-tube, where the nectar is produced and accumulated (Figures 1–3). All the species presented high pollen viability
5mm
Figure 3. 3. Lon Longit gitudi udinal nal sec sectio tion n of flo flower wer of M. pur- pureu pu reus s . Note the nectar nectar chamber (arro (arrow). w). Illust Illustrat ration ion by R. R. Lup Lupo o.
flowering pattern (sensu Newstron et al., 1994). Flowers of M. flaviflorus have outer perianthsegments spreading to erect, reddish pink; the inner perianth-segments opened very slightly and are yellow to pale cream (Figures 4 & 5); those of M. streckeri have outer perianth-segments spreading and deep pink, the inner perianth-segments are erect and purplish pink (Figures 6–9). The flowers of both species are tubular, the stamens are positioned inside the floral tube; the stigma is 4- to 6-lobed, and the stigma-lobes are also inserted (Figures 1–3). Flower length of M. flaviflorus (12 (12.3 .3 ± 0. 0.1m 1mm, m, mean ± SE) was smaller smaller than than that of M. streckeri (13.11 ± 1.2 (13. 1.2mm mm,, mean mean ± SE, t=-2 t=-2.61 .618, 8, P=0 P=0.01 .01,, n=32 for both), whereas the diameter was similar (1.88 ± 0.2 (1. 0.2mm mm for M. flaviflorus , 1.8 1.8 ± 0.9m 0.9mm m for for M. streckeri , mean ± SE, t=-1.593, t=-1.593, P>0.01, n=32 for both, 1). Flowers of M. flaviflorus and M. streckeri started to open at about 7 p.m. (5–10 flowers each individual daily), and lasted c.36 hours, wilting in the morning of the third day. In recently opened flowers the stigma-lobes were closed and the pollen was released; between the end of the first day and the beginning of the second day of the anthesis, the stigma-lobes opened and
became perceptibly moist. The flowers of both species were scentless and produced similar nectar volumes, whereas the concentration of sugar was higher in M. flaviflorus (Table 1). The flowering period of M. purpureus occurred from May to November, and the plants developed fruits between July and December, indicating an annual flowering pattern ( sensu Newstron et al., 1994). The floral tube of M. pur pureus pure us is deep pink-magenta, the outer perianthsegments are recurved and deep pink to reddish; the inner perianth-segments are spreading or recurved, white. The stamens are numerous (c.400), being disposed as a dense ring inside the tube. The style is long, tube-like and exceeds the anthers by about 10mm; the exerted stigma is 10lobed. Flowers are larger than those of the other two species (F=793.64 for corolla length and F=239.81 for corolla diameter, P=0 for both, Table 1). Individuals of Micranthocereus purpureus opened 2-3 flowers each in the afternoon (5 p.m.), which reached full anthesis 2 hour later, remained open during the night, and were completely closed by 10–11 a.m. the next morning. When the flower was completely open, the stigmas were moist (indicating receptivity), laterally or adaxially displaced; after c. 2 hours it began to move slowly towards the center of the flower, where it remained until the end of anthesis. The flowers of M. purpureus presented a slightly sweet and sour scent. Sugar concentration in the nectar decreased during the period between 8 p.m. and 2 a.m. Nectar volume was higher in M. purpureus than in the other two species (F=142.16, P=0), which presented similar volume values (P=0.21); on the other hand, the highest nectar concentration was observed in M. flaviflorus , being followed by M. purpureus and M. streckeri (F=18.29, P<0.05 for all comparisons, Table 1). Pollination process The main floral visitors of Micranthocereus flaviflorus were males of the hummingbird Chlorostilbon aureoventris Boucier & Mulsant (Figure 4), which were observed from early morning until 2 p.m. This hummingbird visited the flowers at regular intervals of approximately 10–30 minutes. The females visited the flowers only sporadically, being frequently chased away by the males during the visits. Chlorostilbon straightt bill (19.09 ± 2.1mm aureoventris has a straigh length, lengt h, mean ± SE, n = 5), that that contacted contacted the the
4
5
6
7
Figure 4. 4. Fem 5. The butt emal alee of of Chlorostilbon aureoventris visiting aureoventris visiting flower of M. flaviflorus . Figure 5. butterfl erfly y Phoebis Figure re 6. 6 . Ma philea phil ea on flowers of Micranthocereus flavifloru fla viflorus s . Figu Male less of of Chlorostilbon Chlorostilbon aureoventris visiting aureoventris visiting flower of 7. Ma M. streckeri . Figure 7. Male less of Chlorostilbon of Chlorostilbon aureoventris on aureoventris on flower of M. streckeri . Not Notee the the prese presence nce of pol pollen len grainss on the bill grain bill of the hummingb hummingbird. ird. Photos by E.R. E.R. Pansa Pansarin. rin.
8
9
10
11
Figure 8. 8. The hummin Figure gure 9. 9. The hesp hummingbird gbird Phaethornis squalidus visiting squalidus visiting flower of M. of M. steckeri . Fi hesperid erid Vettius Vettius sp. sp. Figure gure 10. 10. The ha (arrow) on flowers of M. of M. streckeri . Fi hawkmoth wkmoth Agrius cingulatus visiting cingulatus visiting a flower of M. of M. pu purpu rpureus reus . 11. The hawkm Note the extend proboscis being introduced in the perianth. Figure 11. hawkmoth oth Eryn Erynnyis nyis alope alope visiting visiting a flower of M. of M. purpu purpureus reus . Note the extend extend probo proboscis scis being being introduced introduced in the the perianth. perianth. Photos 8 & 9: 9: E.R. Pansa Pansarin; rin; 10 & 11: L.Y.S .S.. Aon Aona. a.
Table 1. 1 . Flo Floral ral mor morpho phomet metric ricss and and nect nectar ar fea featur tures es (me (mean an ± sd for for both both)) of thr three ee spec species ies of Micranthocereus in Micranthocereus in northeast Brazil. Different letters in the same column indicate statistically different different data. Vol.: vo volum lumee, Con Conc.: c.: con concen centra tratio tion. n. Corolla
Nectar
Species Length (mm)
Diameter (mm)
Vol (µl)
Conc (%)
M. flaviflorus
12.35 12. 35 ± 0.9 0.99 9b
1.76 1. 76 ± 0. 0.25 25b
5.17 5. 17 ± 2. 2.25 25b
23.33 23. 33 ± 1.2 1.28 8b
M. streck streckeri eri
13.09 13. 09 ± 1.2 1.23 3c
1.85 1. 85 ± 0. 0.18 18b
7.3 7. 3 ± 3. 3.8 8b
19.3 19 .3 ± 1. 1.83 83c
M. purpureus
31.88 31. 88 ± 1.8 1.87 7a
16.35 16. 35 ± 1.4 1.49 9a
143.81 143 .81 ± 49. 49.32 32a
21.18 21. 18 ± 1.5 1.54 4a
anthers and the stigma when collecting floral nectar. The visitor prevented access by other individuals of the same species species to the population population of plants which it defended. The butterfly Phoebis philea Johansson (Figure 5) was occasionally observed visiting M. flaviflorus flowers during the hottest hours of the day (12 to 3 p.m.), when it collected nectar and contacted the sexual parts of the flower. The bee Trigona spinipes Fabricius was observed visiting flowers of M. flaviflorus mainly from 10.30 to 11 a.m. and from 2 to 5 p.m., when it collected pollen directly from the anthers, often damaging both anthers, perianth-segments and floral tube, but without contacting the stigma. Flowers of M. streckeri were also pollinated by Chlorostilbon aureoventris (Figures 6 & 7). In this species the females were more frequent as pollinators than males. The visits occurred between 6 a.m. and 4 p.m., being more frequent in the morning and occuring in intervals of approximately 30 minutes, rarely up to 60 minutes. In the afternoon the visits were rare and took place at intervals of 1hour 10 minutes, approximately. Females of C. aureoventris presented the same territorialist behaviour observed in males in M. flaviflorus , mostly during the morning, when they defended the population from males. The hummingbird Phaethornis squalidus Temminck was sporadically observed pollinating M. streckeri flowers in late afternoon (Figure 8), in a similar frequency to that of Chlorostilbon aureoventris (t=0.896, P=0.394, Figure 12); birds of this species were also chased away by females of C. aureoventris . All hummingbird species had the pollen grains deposited on the bill. In M. flaviflorus and M. streckeri , the visits by both species of hummingbird to each flower were brief (c. 2 seconds). In most visits both hummingbirds visit-
ed more than one flower of each plant. Some insects, like the skipper Vettius sp. (Lepidoptera: Hesperiidae: Hesperiinae) species (Figure 5), occasionally visited M. streckeri flowers along the day day,, collected nectar and sometimes touched the sexual parts of the flower with their proboscis, being considered occasional pollinators. Honey bees, Apis melifera, melifera, were observed on flowers of M. streckeri mainly in the morning, when they visited more than one flower of each individual, or even flowers of the same cephalium. They landed at the apex of the floral-tube and collected large amounts of pollen directly from the anthers, sometimes causing anther damage, and rarely contacted the stigma, thus acting mainly as pollen robbers. Trigona spinipes was also observed visiting flowers of M. streckeri at the same period of the day and presenting similar behaviour to that observed with M. flaviflorus . Other occasional visitors were some nonidentified ants, beetles and caterpillar species, which were observed cutting (ants) and eating (beetles and caterpillars) floral parts. Flowers of M. purpureus were visited by the bat Lonchophylla mordax Thomas (Phyllostomidae, Glossophaginae), which was observed after dusk until 5 a.m., in intervals of 30–60 minutes, often approaching the plant from the same direction of the cephalium, hovering in front of the flowers and visiting different individuals at each bout. During its visits, this bat introduced its head inside the floral-tube, touching the stigma and receiving the pollen grains on the head, neck and shoulders. Lonchophylla mordax also visited flowers of Encholirium brachypodum L. B. Sm. & Read (Bromeliaceae) in a higher frequency (intervals of 5–30 minutes) than that observed for M. purpureus , but displaying a similar behaviour. The sphingids Agrius cingulata Fabricius
(Figure10),
Erynnyis alope Drury (Figure11), Erynnyis ello L. (Sphingidae) were observed visiting the flowers of M. purpureus , between 9 p.m.
and 4.30 a.m., with a peak between 9 p.m. and 12 a.m. (Figure 13). The hawkmoths approached the flower and introduced the proboscis into the nectar chamber, hovering at variable distances in front of the flower. Photographs of moths foraging at flowers reveal thick deposits of pollen along their proboscis. The hawkmoths sometimes touched the flowers with their forelegs, mainly E. ello, with visits lasting up to three seconds. All hawkmoths displayed a similar visitation pattern, visiting the flowers more or less in the same sequence and at intervals varying from 5–30 minutes, rarely up to 60 minutes. Agrius cingulata was the most commonly observed visitor, with visits lasting up to 15–20 seconds while making circular movements with the proboscis. Frequency of visits of sphingids was greater than t han that of the bats (t = -2.24, P=0.02, Figure 13). During anthesis a large number of small beetles were were found found insi inside de the flowe flowers rs of M. pur pureus pure us , remaining there even after flower senescence. They were never seen flying from flower to flower or touching the stigma. It was not possible to observe what kind of floral resource these insects explored. The flowers of M. purpureus were also visited by the hummingbirds Chlorostilbon aureoventris in the late afternoon of the first day and Augastes lumachellu lumachellus s Lesson in the early morning of the following day. However, the hummingbird visits were rare and it was not possible to observe whether contact of anthers or stigmas occurred. Apis melifera and Trigona spinipes were observed between 5–8 a.m. and 4 and 5 p.m. (Figure 13), respectively, displaying the same behaviour as was observed with M. flaviflorus and M. streckeri (Figure 12). Discussion Most floral features of M. streckeri and M. flaviflorus are similar to those displayed by species pollinated by hummingbirds in tropical communities, such as the easily accessible and unscented flowers, with tubular and conspicuously coloured perianth, clear separation between floral resource (nectar) and sexual parts and nectar with a relatively low concentration of sugars (Faegri & van der Pijl, 1980; Arzimendi & Ornelas, 1990; Araujo, 1996, Sazima et al .,., 1996; Locatelli & Machado, 1999b; Buzatto et al .,., 2000; Castro & Oliveira, 2001, 2002; Castro & Araujo, 2004), including Cactaceae species (Scogin, 1985).
The short-tubed flowers of M. flaviflorus and M. streckeri allow the visits of both short- and long-billed visitors, like Chlorostilbon aureoventris and Phaethornis squalidus , respectively (Grantsau, 1989) however only M. streckeri was visited by both species. The absence of P. squalidus on M. flaviflorus flowers may be related to the lack of forest habitats near the studied population, since this species is a typical pollinator in Atlantic forests in Brazil (Buzato et al., 2000). This hummingbird seemed to prefer individuals of M. streckeri located in areas at lower elevations, which are close to a wooded area. The period in which P. squalidus and Trigona spinipes visited M. streckeri flowers (after 2 p.m.), and that in which T. spinipes visited M. flaviflorus flowers coincided with the decrease in the number of C. aureoventris visits, a strategy that probably minimizes the chances of being attacked by this hummingbird. It was observed in the populations of Micranthocereus streckeri that Phaethornis squalidus invades the territory of Chlorostilbon aureoventris after 11 a.m. (Figure 12). According to Buzato et al. (2000), hermit hummingbirds such as P. squalidus are the major pollinators of hummingbird-pollinated of floras in Brazilian forests. Phaethornis squalidus visited the plants that occur near to the valley, nearby the forested area. This behaviour might be related with the visits decrease of C. aureo aureoventr ventris is after 3 p.m. and the visits increase of P. squalidus . The decrease in the frequency of visits of Chlorostilbon aureoventris to the flowers of M. flaviflorus after 2 p.m. could be explained by the presence of a sympatric Cactaceae species, Melocactus paucispinus Heimen. & R. J. Paul, whose flowers opened approximately at 2 p.m. and were also visited by C. aureoventris . However, it is necessary to test if this hummingbird really prefers M. paucispinus flowers in comflavif viflor lorus us . parison to those of M. fla The visiting behaviour of Chlorostilbon aureoventris to the flowers of M. flaviflorus and M. streckeri revealed two types of feeding strategies: territoriality (sensu Feinsinger, 1978; Stiles, 1981) and traplining. The traplining strategy favours cross-pollination because it maximizes pollen dispersion among different individuals (Janzen, 1971), whereas the territorialist behaviour may reduce pollen flow in a population (Linhart, 1973). The occurrence of these two types of foraging behaviour (territorialist and traplining) was also recorded for C. aureoventris
6
s t i s i V f o r 3 e b m u N
0 8-9
9-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
Time Interval
Figure 12. 12. Fre requ quen ency cy of vi visi siti ting ng bout boutss of M. flaviflorus and M. streckeri pollina pollinators tors in NE Brazil. Brazil. Males of of Chlorostilbon aureoventris on aureoventris on flowers of M. flaviflorus (hatched flaviflorus (hatched bars); Phaethornis squalidus (grey squalidus (grey bars) and male of C. aureoventris (black aureoventris (black bars) on M. streckeri flowers. streckeri flowers.
12
12
s t i 8 s i V f o 6 r e b m 4 u N 2
0 17-18
18-19
19-20
20-21
21-22
22-23
23-24
24-01
01-01
02-03
03-04
04-05
05-06
Time Interval
Figure 13. 13. Fre requ quen ency cy of of vi visi siti ting ng bou bouts ts of of M. pu purpur rpureus eus pollin pollinator atorss in NE Brazil Brazil.. Sphin Sphingids gids (hatched (hatched bars), bars), Loncophylla mordax (white mordax (white bars), Chlorostilbon aureoventris (black aureoventris (black bars) and Augastes lumachellus (grey lumachellus (grey bars).
in the flowers of Melocactus zehntneri (Britton & Rose) Luetzelb. and Opuntia palmadora Britton & Rose by Locatelli & Machado (1999b). The protandry observed in M. flaviflorus and M. streckeri is considered a reproductive strategy that minimizes the interference between male and female functions as well as the chances of self-pollination (Webb & Lloyd, 1986). Protandry was not found in M. purpureus , where the position of the stigma (adaxial or lateral) in relation to the stamens changed from adaxial or lateral to central in the flower’s lifespan. Flowers of M. purpureus present some typical
chiropterophilous features such as pale colour, a short floral tube with wide aperture, nocturnal anthesis and large amounts of nectar (Vogel, 1969; Faegri & van der Pijl, 1980). There is evidence that the presence of a lateral sunken cephalium, composed by congested, hairy flowerbearing areoles, forming a cushion-like structure is related with the protection of bat’s wings during pollination (Zappi, 1994). The exerted position of the stigma of M. purpureus seems to favour cross-pollination because it is the first reproductive part to be touched by the floral visitors (Sazima & Sazima, 1975). However, a
relatively low frequency of visits by bats was recorded. The horizontally disposed position of the flowers, as well as its presence at only on one side of the plants of M. purpureus plants are features expected in chiropterophilous species (Vogel, 1969, Sazima et al .,., 1995; Machado et al., 1998), because it seems that the one-side presentation of the flowers facilitates their access by bats, which are less able to maneuver in flight than bees or hummingbirds. The low number of flowers and sweet scent presented during the anthesis by M. purpureus was found in bat-pollinated species in general (Sazima & Sazima, 1978; Machado et al., 1998; Sazima et al., 1999) as well as in other bat-pollinated Cactaceae species (Locatelli et al., 1997). The regular paths coupled with the intervals between visiting bouts observed for Lonchophylla mordax visits to M. purpureus indicate a trapline behaviour, as has been recorded for Glossophagine species when feeding on flowers (Baker,, 1973; Fleming, 1982; Sazima et al .,., 1999). (Baker This foraging pattern was already recorded in several studies on bat-pollinated flowers in the Brazilian Atlantic rainforest (Sazima et al., 1994, Machado et al., 1998; Sazima et al., 1999), “restinga” (Locatelli et al .,., 1997) and for L. boker., 1989). manii in “campo rupestre” (Sazima et al ., Encholirium rium brach brachypoypoThe flowers of bromeliad Encholi dum are in the great number (c.30 flowers/inflorescence daily) than M. purpureus and have disagreeable odour, more typical of bat-pollinated species. These characters observed in E. brachy podium at the study site may partially explain the preference of the L. mordax for the bromeliad and the irregular intervals in feeding bouts on M. purpureus purpur eus . Despite the hawkmoths being generally considered less efficient and less important pollinators than bats in chiropterophilous species ( sensu Baker et al., 1971; Sazima & Sazima, 1978; Sazima et al .,., 1994), these insects were more frequently observed visiting M. purpureus flowers than bats. These results indicate that M. pur pureus pure us may have, at the sudy site, a mixed bathawkmoth pollination system, similarly to the one described for Lafoensia densiflora Pohl (Silberbauer-Gottsberger & Gottsberger, 1975) and for Cereus peruvianus auctt. [=C. hildmannianus K. Schum.] (Silva & Sazima, 1995). In fact, M. purpureus seems to present floral features such as the flower odour, which is weak and sweet, rather than strong and disagreeable that are somewhat ambiguous between bat and hawk-
moth syndromes, suggesting an adaptation to attract both polli pollinator nator types. Moreov Moreover er,, the the traplining strategy presented by many hawkmoth species (Janzen, 1971), that was also observed in this study, coupled with their quick movements and long flight distances (Linhart & Mendenhall, 1977), favour cross-pollination. Despite the frequent presence of beetles within the flowers of Micranthocereus purpureus , similarly observed in other species of Cactaceae (Silva & Sazima, 1995; Locatelli et al., 1997), some studies suggested that these insects are not important pollen vectors for most cacti species (Grant & Hurt, 1979), but seem to act only as robbers (Locatelli et al., 1997). The hummingbirds and bees that were observed contacting anthers and stigma during their visits to M. pur pureus pure us flowers should not be considered pollinators because no stigma receptivity was observed in these periods of anthesis (Figure 13). Moreover, the flowers had nearly empty anthers when visited by Augastes lumachel lumachellus lus Lesson, which was observed only once; this Trochilidae species was observed using M. purpureus hairs for building its nest (Machado et al., 2003). The potential for using both nocturnal and diurnal pollinators is known in Cactaceae (McGregor et al.., 1962) and in other chiropterophilous al angiosperms (Buzato et al., 1994; Sazima et al .,., 1994). The range of nectar concentration of M. pur pureus was slightly higher than the values recorded for some Neotropical bat-pollinated species (5%–29%, often 15%–17%, von Helversen, 1993; Sazima et al., 1999). Scogin (1985) compared nectar analyses in some Cactaceae species as 22–30%, 15–34% and 17–26%, for those respectively pollinated by hummingbirds, moths and bats. The value found for M. purpureus was similar to that found in other Brazilian bat-pollinated Cactaceae (Locatelli et al .,., 1997) and in species of other families in Brazil (Sazima et al., 2003). The present study found overlapping values of sugar concentration between Micranthocereus , with the two hummingbird pollinated species (M. flaviflorus and M. streckeri ) at each range of the spectrum, while the bat pollinated M. purpureus had intermediate values between the two (Table 1). It was also found that the range of variation in sugar concentrations in Micranthocereus may favour relationships with less specific pollinators than was initially expected. Tribe Cereeae (Taylor & Zappi, 2004) comprises several closely
related genera, each one with more than one pollinator and sometimes displaying gradients between pollination syndromes. For example, Stephanocereus and Coleocephalocereus are likely to be pollinated by bats and hummingbirds, and species of Pilosocereus are pollinated by bats and hawkmoths was studied by Locatelli et al. (1997), Locatelli & Machado (1999a) and E. A. Rocha, pers. com.). Our results show a remarkable variability in the flower biology within the genus Micranthocereus . Ac k no w le dg e me nt s We thank Dr. Nigel Taylor and Dr. Volker Bittrich for kindly read the manuscript, Dr. Dr. Keith J. Brown for identification of the species sphingids, Dr. Ivan Sazima and Dr. Marlies Sazima for identification of the bat species, Ludmila Mickeliunas and Renato P. Oliveira for their assistance in the field studies and Rogério Lupo for the illustrations. This study is part of a Master thesis by the first author for the postgraduation course in Plant Biology at the ‘Universidade Estadual de Campinas’ (São Paulo, Brazil). This work was supported by the Research Support Foundation of São Paulo State - FAPESP FAPESP (process no. 00/00369-7). References A ONA (2003) 03).. Carac Caracter teriza ização ção e ONA , L. Y. S. (20 delimitação do gênero Mic ranthocereus Backeb. (Cactaceae) baseados em caracteres morfológicos e moleculares . M. S. Thesis,
Universidade Estadual de Campinas, SP. A RAÚJO C. (19 (1996). 96). Beija-flores e seus recursos RAÚJO, A. C. florais numa área de planície costeira do litoral norte de São Paulo . M. S. Thesis,
Universidade Estadual de Campinas, SP. A RZIMEDI (199 990) 0).. RZIMEDI, M. C. & O RNELAS, J. F. (1 Hummingbirds and their flower resources in 22:: a tropical dry forest in Mexico. Biotropica 22 172–180. B AKER, H. G., CRUDEN, R. W. & B AKER, I. I. (19 (1971 71). ). Minor parasitism in pollination biology and its community function: the case of Ceiba 21:: 1127–1129. acuminata . BioScience 21 B AKER, H. G. (1973). Evolut Evolutionar ionaryy relationshi relationships ps between flowering plants and animals in American Americ an and Afric Africaa trop tropical ical fores forests. ts. In Meggers, B. J., Ayensu, E. S., & Duckworth, W. D. (Eds.), Tropical forest ecosystems in Africa and South America: A comparative Smithsonian Institute Press, review .
Washington, DC., pp. 145–159.
B ARTHLOTT, W. & H UNT, D. (1993). Cactaceae. In Kubitzki, K., Rohwer Rohwer L. & Bittrich, V. V. (Eds.), The families and genera of vascular plants , Vol. V ol. II. Springer, Springer, Berlin, pp. 161–197. BIANCHI, M. B., GIBBS, P. E., P RADO, D. E. & VESPERINI, J. L. (2000). Studies on the breeding systems of understorey species of a Chaco 1955: woodland in NE Argentina. Flora 19 339–348. BOYLE, T. H. (1996). Characteristics of self-incompatibility in Schlumbergera truncata and S. 53. buckleyi (Cactaceae). Sex. Pl. Repro. 9: 49 – 53. BUZATO, S., S AZIMA , M. & S AZIMA , I. (1994). Pollination of three species of Abut Abutilon ilon (Malvaceae) intermediate between bat and 189:: hummingbird flowers syndromes. Flora 189 227 – 334. 334. BUZATO, S., S AZIMA , M. & S AZIMA , I. (2000). Hummingbird-pollinated floras at three 32:: 824–841. Atlantic forest sites. Biotropica 32 C ASTR O, C. C. & O LIVEIRA , P. E. (2001). Reproductive biology of the protandrous Fernandusa speciosa Pohl (Rubiaceae) in 24:: southeastern Brazil. Rev. Bras. Bot. 24 167–172. C ASTR O, C. C. & O LIVEIRA , P. E. (2002). Pollination biology of distylous Rubiaceae in the Atlantic rain forest, SE Brazil. Plant Biol. 5: 640–646. C ASTRO, C. C. & A RAUJO A.C. (2004). (2004). Disty Distyly ly and RAUJO, A.C. sequential pollinators of Psychotria nuda (Rubiaceae) in the Atlantic rain forest, Brazil. 244:: 131–139. Plant Syst. Evol. 244 FEINSINGER, P. (1978). Community organization among neotropical nectar-feeding birds. Amer. 18:: 779–795. Zool. 18 F AEGRI, K. & V AN DER PIJL, L. (1 (198 980) 0).. The princi ples of pollination ecology ecology.. 3rd ed. (Reprint with revisions), Pergamon Press, New York. FLEMING, T. T. H. (1982). Forag Foraging ing strateg strategies ies in in plant-visiting bats. In: Kunz, T. H. (Ed.), Ecology of bats . Plenum Press, New York., pp. 287–325. FLEMING, T. H., TUTLE, M. D. & H ORNER, M. A. (1996). Pollination biology and the relative importance of nocturnal and diurnal pollinators in three species of Sonoran Desert colum41:: 257–269. nar cacti. Southwest. Nat . 41 GRANT, K. A. & H URT, P. P. D. (1979). Pollin Pollination ation of the southwestern American opuntias. Pl. 133:: 15–28. Syst. Evol. 133 GRANTSAU, R. (1989). Os beija-flores do Brasil . Expressão e Cultura, Rio de Janeiro.
H ARBER, W. A. & F RANKIE, G. W. W. (1989). A tropical community: Costa Rican community dry 21:: 155–172. forest Sphingidae. Biotropica 21 HOLLAND , J. N.& F LEMING, T. H. (1999). Mutualistic interactions between Upiga virescens (Pyralidae), a pollinating seed-consumer, and Lophocereus schottii (Cactaceae). 80:: 2074–2084. Ecology 80 HUNT, D. R. & T AYLO AYLOR R, N. P. (1990). The genera of Cactaceae: progress towards consensus. Bradleya 8: 85–107. J ANZEN, D. H. (1971). Euglossine bees as long-distance pollinators of tropical plants. Science 171:: 203–205. 171 LINHART, Y. S. (1973). Ecological and behavioral determinants of pollen dispersal in hummin1077: bird-pollinated Heliconia. Am. Nat . 10 511–525. LINHART, Y. S. & MENDENHALL, J. A. (1977). Pollen Pollen dispersal by hawkmoth in Lindenia rivalis Benth. population in Beliza. Biotropica 9: 143. LOCATELLI, E., M ACHADO, I. C. S. & M EDEIROS, P. (1997). Floral biology and pollination in Pilosocereus catingicola (Cactaceae) in 15:: 28–34. Northeastern Brazil. Bradleya 15 LOCATELLI, E. & M ACHADO, I. C. S. (1999a). Floral biology of Cereus fernambucensis : a sphin17:: gophilous cactus of restinga. Bradleya 17 86–94. LOCATELLI, E. & M ACHADO, I. C. S. (1999b). Camparative study of the floral biology in two ornithophilous species of Cactaceae: Melocactus zehntneri Opuntia and 17:: 75–85. palmadora palmad ora . Bradleya 17 M ACHADO, C. G., M OREIRA , T. A., N UNES, C. E. C. & ROMÃO, C. O. (2003). Use of Micranthocereus purpureus (Guerke) F. Ritter (Cactaceae) hairs in nests of Augastes lumachellus Lesson (Trochilidae, Aves). Sitientibus sér. Ciên. Biol. 3: 131–132. M ACHADO, I. C. S., S AZIMA , I. & S AZIMA , M. (1998). Bat pollination of the terrestrial herb Irlbachia alata (Gentianaceae) in northeast209:: 231–237. ern Brazil. Pl. Syst. Evol . 209 M ACHADO, M. C. (1999). The cacti of Morro do 17:: Chapéu, Bahia, Brazil. Brit. Cact. Succ. J. 17 203–213. MCF ARLAND, J. D., K EVAN EVAN, P. G. & L ANE, M. A. (1989). Pollination biology of Opuntia imbricata (Cactaceae) in southern Colorado (USA). 67:: 24–28. Can. J. Bot. 67 M ANDUJANO, M. D. C, M ONTANA , C., MENDEZ, I. & GOLUBOV, J. (1998). The relative contributions contributions
of sexual reproduction and clonal propagation in Opuntia rastrera from two habitats in the 86:: 911–921. Chihuahuan Desert. J. Ecol. 86 MCGREGOR, S. E., A LCORN LCORN, S. M. & O LIN, G. (1962). Pollination and pollinating agents of 43:: 259–267. the saguaro. Ecology 43 METZ, C., NERD, A. & MIZRAHI, Y. (2000). Viability of pollen of two fruit crop cacti of the genus Hylocereus is affected by temperature and 35:: 199–201. duration of storage. Hortscience 35 NEGRON-ORTIZ, V V.. (1998). Reproductive biology of a rare cactus, Opuntia spinosissima (Cactaceae), in the Florida Keys: Why is seed 11:: 208–212. set very low? Sex. Pl. Repro. 11 NEWSTRON, L. E., F RANKIE, G. W. & B AKER, H. G. (1994). A new classification for plant phenolophenology based on flowering patterns in lowland tropical rain forest trees at La Selva, Costa 26:: 141–159. Rica. Biotropica 26 PORSCH, O. (1939). Cactaceae. In: Druck, J. (Ed.), Das Bestäubungsleben der Kakteen-blüte, II . 1939:: 81–142. Cactaceae (DKG) 1939 R ADFORD, A. E., D ICKINSON, W. C., M ASSEY , J. R. & BELL, C. R. (1974). Vascular plant systematics . Harper & Tow Publishing, New York. R AW, A. (1996). Territories Territories of the t he ruby-topaz hummingbird Chrysolampis mosquitus at flowers of the “turk’s-cap” cactus, Melocactus salvadorensis in the dry caatinga of the north56:: 581–584. Rev.. Bras. Biol . 56 eastern Brazil. Rev RUIZ, A., S ANTOS, M., SORIANO, P. J., C A AVELIER VELIER, J. & C ADENA , A. (1997). Mutualistic relations between the bat Glossophaga longirostris and the columnar cacti in the arid region of La 29:: 469–479. Tatacoa, Colombia. Biotropica 29 RUSCHI, A. (1986). Aves do Brasil, Beija-flores . Expressão e Cultura, Rio de Janeiro. S AHLEY , C. T. (2001). Vertebrate pollination, fruit production, and pollen dispersal of Stenocereus thurberi (Cactaceae). Southwest. 46:: 261–271. Nat. 46 S AZIMA , I., VOGEL, S. & S AZIMA , M. (1989). Bat pollination of Encholirium glaziovii , a terres158:: 167–179. trial bromeliad. Pl. Syst. Evol . 158 S AZIMA , I., B UZATO, S. & S AZIMA , M. (1996). An assemblage of hummingbird-pollinated flowers in a montane forest in southeastern 109:: 149–160. Brazil. Bot. Acta 109 S AZIMA , M. & S AZIMA I. (1975). Quiropterofilia em Lafoensia pacari St. Hil. (Lythraceae), na 27:: Serra do Cipó, Minas Gerais. Ciên. Cult. 27 404–416. S AZIMA , M. & S AZIMA , I. (1978). Bat pollination of the Passion flower, Passiflora mucronata, in
10:: 100–109. Southeastern Brazil. Biotropica 10 S AZIMA , M., S AZIMA , I. & BUZATO, S. (1994). Nectar by day and night: Siphocampylus sulfureus (Lobeliaceae) pollinated by humming191:: 237–146. birds and bats. Pl. Syst. Evol. 191 S AZIMA , M., BUZATO, S. & S AZIMA , I. (1995). Bat pollination of Vrisea in southeastern Brazil. Bromélia 2: 29–37. S AZIMA , M., BUZATO, S. & S AZIMA , I. (1999). Batpollinated flower assemblages and bat visitors at two Atlantic forest sites in Brazil. Ann. 83:: 705–712. Bot . 83 S AZIMA , M., BUZATO, S. & S AZIMA , I. (2003). Dyssochroma viridiflorum (Solanaceae): a reproductively bat-dependent epiphyte from the Atlantic Rainforest in Brazil. Ann. Bot . 92:: 725–730. 92 SCHLINDWEIN , C. & W ITTMANN, D. (1995). Specialized solitary bees as effective pollinators of south Brazilian species of Notocactus and Gymnocalycium (Cactaceae). Bradleya 13:: 25–34. 13 SCHLINDWEIN, C. & WITTMANN, D. (1997). Stamen movements in flowers of Opuntia (Cactaceae) favour oligolectic pollinators. Pl. Syst. Evol . 204:: 179–193. 204 SCOBELL, S. A. & SCOTT, P. E. (2002). Visitors and floral traits of a hummingbird-adapted cactus (Echinocereus coccineus ) show only minor variation along an elevational gradient. Am. 147:: 1–15. Mid. Nat . 147 SCOGIN, R. (1985). Nectar constituents of the 30:: 77–82. Cactaceae. Southwest. Nat . 30 SILVA , W. R. & S AZIMA , M. (1995). Hawkmoth pollination in Cereus peruvianus , a columnar cactus from southeastern Brazil. Flora 190:339–343. 190 :339–343. SILBERBAUER-GOTTSBERGER, I. & G OTTSBERGER, G. (1975). Über Shingophile Angiospermen 123:: 157–184. Brasiliens. Plant Syst. Evol . 123 SOKAL, R. R., R OHLF, J. R. F. (1969). Biometry: the principles and practice of statistics in biological research. W. H. Freeman, San Francisco.
STILES, F. F. G. (1981). Geographical aspects of birdflower coevolution, with particular reference 68:: to Central America. Ann. Miss. Bot. Gard . 68 323–351. SUZAN, H., N ABHAN, G. G. P & P ATTEN, D. T. (1994). Nurse plant and floral biology of a rare nightblooming cereus Peniocereus striatus (Brandegee) F. Buxbaum. Conserv. Biol . 8: 461–470. T AYLO AYLOR R, N. P. & Z APPI, D. (2004). Cacti of Eastern Brazil . Royal Botanic Gardens, Kew. TSCHAPKA , M., VON HELVERSEN, O. & B ARTHLOTT, W. (1999). Bat pollination of Weberocereus tunilla , an epiphytic rain forest cactus with functional flagelliflory. Plant Biol. 1: 554 – 559. 559. VIANA , M. L., O RTEGA , B. P., S ARA ARAVIA VIA , M., B AO, E. I & SSCHLUMPBERGER, B. (2001). Biologia floral y polinizadores de Trichocereus pasacana (Cactaceae) en el Parque Nacional Los 49:: Cardones, Argentina. Revist. Biol. Trop . 49 279 – 285. 285. VOGEL, S. (1969). Chir Chiropter opterophil ophilie ie in der neotropischen Flora. Neue Mitteilungen II. Flora 158:: 185 – 222. 158 222. VOGEL, S. (1990). Radiacion adaptativa del sindrome floral en las familias neotropicais. Bol. 59:: 4 – 30. Acad. Nac. Cien. Cien. Córdoba Argentina 59 30. VON HELVERSEN, O. (1993). Adaptations of flowers to pollinationd by Glossophagine bats. In Barthlott, W. Naumann, C. M., SchimidtLoske, K. & Schumann, K. L. (Eds.), AnimalPlant interactions in tropical environments . Museum Alexandre König, König, Bonn, pp. 41–59. W ALLACE, R. S. & C OTA , J. H. (1996). An intron loss in the chloroplast gene rpo Ci Ci supports a monophyletic origin for the subfamily 29:: 275–281. Cactoideae. Curr. Genet. 29 WEBB, C. J., L LOYD, D. G. (1986). The avoidance of interference between the presentation of pollen and stigmas in Angiosperms. I. 24:: 135 – 162. Dichogamy. New Zeal. J. Bot. 24 162. Z APPI, D. C. (1994). Pilosocereus (Cactaceae): The genus in Brazil. Succ. Pl. Res . 3: 1–160.
