| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biology, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5E2
Received for publication December 6, 2001. Accepted for publication May 7, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Apiaceae • Carum carvi • dichogamy • flower phenology • nectar carbohydrates • nectar-sugar reabsorption • protandry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Vogel, 1983
; Shuel, 1992
). Owing to its significant roles in attracting potential pollinators and in serving as the precursor of honey, floral nectar understandably has received widespread research attention. Compared to other items such as pollen and oils, the production of nectar is a relatively inexpensive investment for plants to attract flower visitors (Simpson and Neff, 1983
). Although estimated expenditures for nectar production have ranged from 4.3 to 36.6% of daily photosynthate in Asclepias syriaca (Southwick, 1984
), 30% of the energy devoted to flowering in A. quadrifolia (Pleasants and Chaplin, 1983
), and 20% of the energy contained in harvested foliage of Medicago sativa (Southwick, 1984
), a more recent analysis said to represent several bee-pollinated angiosperms estimates that floral nectar comprises only 3.3% of the energy content per flower of Pontederia cordata (Harder and Barrett, 1992
). Indeed, another characteristic feature of nectar is its ability to be reabsorbed by the plant (Bonnier, 1879
; Pedersen, LeFevre, and Wiebe, 1958
), even during the period of active secretion (Ziegler and Lüttge, 1959
; Shuel, 1961
; Bieleski and Redgwell, 1980
; Nicolson, 1995
). Provided this resource maintains contact with the persistent floral nectary, ungathered nectar may eventually be reclaimed (Bonnier, 1879
; Búrquez and Corbet, 1991
; Davis, 1997
).
There are two nectaries within each small, nontubular floret of the Apiaceae; each gland is usually located at the base of the stylopodium atop an inferior ovary and secretes its nectar through surface stomata (Behrens, 1879
; Magin, 1983
). The numerous compound inflorescences (umbels) per plant, consisting of florets arranged in smaller umbels (umbellets), create visually prominent landing platforms of closely aggregated florets ("pseudanthia"; Weberling, 1992
) with readily available nectar and pollen attractive to bees, flies, and other insects (Proctor, Yeo, and Lack, 1996
). Potential honey production from species of Apiaceae is estimated at 27.8 kg/ha for Anethum graveolens (Warakomska, Kolasa, and Wróblewska, 1982
), 134–253 kg/ha for Angelica silvestris (Jablonski and Koltowski, 1993
), 62.4–160 kg/ha for Petroselinum sativum (Warakomska, Kolasa, and Wróblewska, 1983
), and 70–134 kg/ha for caraway (Carum carvi L.; Langenberger and Davis, 2002
).
Caraway, one of the oldest herbs known (Németh, 1998
; Levetin and McMahon, 1999
), is native to Asia, Europe, and North Africa (Rosengarten, 1969
; Levetin and McMahon, 1999
). It is utilized traditionally as a condiment, oil, and drug, and more recently for carvone, which inhibits sprouting in potatoes (Németh, 1998
; Putievsky, 1998
). Production of caraway as a spice crop has been increasing in the northwestern United States (Rosengarten, 1969
; McGregor, 1976
) and Canada (Anonymous, 2000
), where it is now widely escaped from cultivation (Rosengarten, 1969
; Mihalik, 1998
). In terms of reproductive structures and strategy, each compound umbel is arranged in a hierarchy of increasing umbel orders (1°, 2°, 3°, and so on), the next higher-order umbel arising in the axil of the bract below the umbel presently flowering (Rosengarten, 1969
, p. 148). Umbels per plant can reach 5° order subject to environmental conditions (Langenberger and Davis, 2002
). Caraway is andromonoecious, with its perfect florets being protandrous and strongly dichogamous (Knuth, 1908
), the stigmata not being receptive until several days after pollen is shed (Van Roon and Bleijenberg, 1964
; Bouwmeester and Smid, 1995
). The outer, perfect florets in an umbellet open first, whereas the inner florets, usually staminate, open next (Bouwmeester and Smid, 1995
). All umbels of the same order per plant usually flower simultaneously, such that only in exceptional circumstances does self-pollination in this self-compatible species occur within the umbel (Németh, Bernáth, and Petheô, 1999
; Németh and Székely, 2000
). However, umbels of the next order begin to flower while the previous order still has umbellets with female-phase florets, allowing for geitonogamy to occur (Németh, Bernáth, and Petheô, 1999
).
As a result of the marked dichogamy within a perfect floret of caraway, phenology passes through distinct sexual phases, a male phase of anther dehiscence and stamen loss followed up to 7 d later by a female one during which style elongation, stigma receptivity, and pollination occur (Van Roon and Bleijenberg, 1964
; Bouwmeester and Smid, 1995
). Floral nectar is produced during both phases in Coriandrum sativum and Daucus carota (Koul, Hamal, and Gupta, 1989
; Koul, Sharma, and Koul, 1993
). However, information about nectar production throughout floret phenology in caraway is lacking, and to our knowledge there are no previous studies in the Apiaceae for which extensive data on nectar are available for the separate male and female phases of the same florets. Indeed, given the paucity of nectar data in this family and the certain biological importance of exposed nectar of Apiaceae to a wide variety of short-tongued insect taxa, including those useful as parasites and predators in biological control (Proctor, Yeo, and Lack, 1996
), this detailed study of caraway was undertaken.
Accordingly, our objectives were to examine nectar dynamics as well as changes in composition of major carbohydrates during the two distinct sexual phases that encompass the lengthy period of flowering in annual caraway. As no nectar data are available for different genotypes of caraway, two varieties were investigated. Any influence of umbel order on nectar production was also studied. Moreover, upon the discovery of a regular period of complete nectar reabsorption and absence of nectar secretion between a floret's sexual phases, experiments were performed to investigate the effect of nectar removal during the male phase on total nectar production per floret.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
For 12 consecutive days in March, 1998, six newly opened perfect florets from the same representative greenhouse plant were tagged. On the final day, the florets of different ages were simultaneously collected and fixed overnight in 2% glutaraldehyde, 1% fructose in 25 mmol/L Na phosphate buffer (pH 6.8) at 4°C. After three washes with buffer, tissues were post-fixed for 2 h in 1% OsO4 in buffer. Following three rinses with buffer and then distilled water, tissues were dehydrated in a graded acetone series before critical-point drying (Polaron Instruments, Watford, UK). Florets were mounted using double-sided sticky tape on stubs and gold-coated with an Edwards Sputter Coater S150B before viewing at 30 kV with a Philips 505 scanning electron microscope. Photographs taken with Polaroid 665 positive/negative film were printed on Kodak Polycontrast III RC paper.
Plants in a growth chamber
To restrict environmental fluctuations during growth and the following experiments involving nectar collection, two varieties of annual caraway (Karzo, Moran) were sown, watered, thinned, and staked (as described above) in a growth chamber. Plants grew in 15 h light and 9 h darkness at 26.5°C and 14°C, respectively. Incandescent and fluorescent lights gave intensities ranging from 220 to 335 µmol·m–2·s–1 at the pot height to plant top, respectively.
Daily floral observations and nectar collection
Three plants of var. Moran (designated A–C) and five of var. Karzo (D–H) were studied in the growth chamber for daily nectar production and flowering progression. For standardization purposes, all central florets (mostly staminate) were removed using fine forceps from experimental umbellets, leaving ten perfect florets nearing anthesis per umbellet, each of which was located on a different 2° or 3° umbel. Four to eight umbellets treated in this way per plant were tagged. Each day, starting at anthesis and continuing until all florets had dried up, observations were made on floret phenology and measurements taken of nectar production, in each experimental umbellet.
Nectar from each perfect floret per umbellet was collected at the same time (9 h after photoperiod began) each day and pooled using a 1-µL capillary tube (Microcaps, Drummond Scientific, Broomall, Pennsylvania, USA). Nectar volume was determined by measuring height of the nectar column in the common-bore tube, and its solute concentration taken at once using a refractometer (40–85%; Bellingham & Stanley, Tunbridge Wells, Kent, UK) modified by the manufacturer to accommodate small volumes. Concentrations were corrected to 20.0°C and expressed as grams per 100 mL using the formula of Búrquez and Corbet (1991)
. For standardization in reporting these data, all nectar volumes were adjusted to the volume corresponding to the average solute concentration. Total nectar production from each plant, and within each sexual phase per floret, was summed and compared using two-tailed t tests (
= 0.05).
Eleven stages (a–k) of floret phenology were designated as listed in Table 1. The one or two most representative stages were assigned for each experimental umbellet per day. For example, if eight florets were at Stage g and two at Stage h, then Stage g was reported, even though nectar was collected from the two Stage-h florets. However, if seven florets were at Stage g and three at Stage h, then Stage gh was assigned. If three stages were represented almost equally, then the intermediate stage was registered.
|
Effect of male-phase nectar collection on female-phase nectar production
On each of four plants (F–I) of var. Karzo in the growth chamber, five pairs of matching umbellets (each pair on its own 2° umbel) were set up and nectar sampled exactly as described earlier for daily floral-nectar collection, except that one umbellet in each of the five pairs was sampled during the stage of stigmatic receptivity (throughout Stage h) only. That is, any nectar produced during the male phase (Stages b–d) was not collected from the latter, female-phase-only umbellets, but allowed to be reabsorbed by the florets. Nectar production in the pairs of experimental umbellets from each plant was compared using a one-tailed t test to determine whether more nectar was produced during the female phase by umbellets whose male-phase nectar was reabsorbed. These results were also compared to the combined (total) quantity of nectar collected from both phases of the other experimental umbellets from each plant, using one-tailed t tests ( = 0.05).
Nectar-carbohydrate composition
On the same day, 5 h after the photoperiod began, nectar from each sexual phase from one plant per variety was collected separately in 1-µL Microcaps from florets in 4° umbels (female phase: Stage h) or 5° umbels (male phase: Stages b–d). Approximately 12 florets in the female phase and 30–40 florets in the male phase were required to fill each microcapillary. Nectar was then expelled onto separate filter-paper wicks, which were allowed to dry and then stored in labeled envelopes until analysis of major nectar carbohydrates (fructose, glucose, and sucrose; other oligosaccharides are rare in the family; Percival, 1961
). After elution of each wick's nectar in 1.0 mL of ultra-distilled water (3°C), samples were filtered, diluted, and analyzed by high performance liquid chromatography (HPLC) (Dionex Bio LC 4000; Sunnyvale, California, USA) by Mr. K. Lew and Prof. N. H. Low, Department of Applied Microbiology and Food Science, and compared to known standards as described in Davis et al. (1994)
, with the following exceptions. Carbohydrates were detected by a Dionex pulsed amperometric detector with a single gold electrode maintained at the following potentials and durations: E1 = 0.05 V (T1 = 0.120 s); E2 = 0.60 V (T2 = 0.120 s); E3 = –0.80 V (T3 = 0.420 s). Each sample was analyzed in duplicate, and the reported values are means.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
, p. 148) corresponded with Stage a and the early part of Stage b from Tables 1 and 2. Prior to opening, the buds were tightly closed. Buds on the periphery of the umbellets opened first and quickly; florets entered the male phase (anther dehiscence) within a few hours after buds began to open. Central florets on each umbellet lagged about 2 d behind.
Male phase
This phase spanned the latter part of Stage b, Stage c (Fig. 3 in Rosengarten, 1969
) and the early part of Stage d (Tables 1 and 2). In young, newly opened florets, five stamens were evident and the two styles were short and directed toward each other (Fig. 1). The stamens dehisced one by one, the earliest-splitting anthers pushing out from among the still partly-folded petals alternating with them. By the time all stamens had dehisced, the five identical petals had unfolded and the floret was fully open. Within a few hours, after all stamens had dehisced, the earliest to dehisce began to detach. The male phase usually ended on day 3 but extended to day 4 in many umbellets on plant B (Table 1) and plant G (Table 2). Duration of the male phase was 3.03 ± 0.03 d (mean ± 1 SE; N = 40 umbellets) and 3.60 ± 0.11 d (N = 20) for the Karzo and Moran varieties, respectively (P < 0.001).
|
Style elongation
This initial stage of a floret's female phase corresponded to Stages f and g (figs. 8 and 7, respectively, in Rosengarten, 1969
), which typically occurred on the fourth or fifth days (Tables 1 and 2). Before elongation, the styles criss-crossed (Fig. 2), with the stigmata facing inward toward the center of the floret. Thereafter, the styles separated, faced away from each other, and continued to elongate for slightly over a day, nearly attaining their full length (Fig. 3).
Style elongation began at about the same time on all perfect florets on an individual umbellet, regardless of how long the florets had been open. Sometimes within the perfect florets near the umbellet's interior, initiation of style elongation preceded the completed dehiscence of all stamens. Stamen loss always occurred before style elongation was complete, but sometimes only marginally. For instance, in umbellets F-8 and G-1 (Table 2), Stages f and g passed rapidly. Overall, within a floret the male phase was always complete before the female phase commenced, but this pattern rarely occurred in synchrony throughout all florets of an umbellet and never was manifested at the level of the umbel.
Prior to the stage of style elongation, the staminate and perfect florets looked similar, but then could easily be distinguished because the styles remained vestigial in the former.
Stigmatic receptivity
This phase began once the styles had fully elongated, corresponding to Stages h and i from Tables 1 and 2. At this point, the floret was receptive to pollination; the stigmatic papillae had separated and a sticky secretion, presumably aiding the entrapment and hydration of pollen grains, appeared. This stage of stigmatic receptivity lasted longer than any other in floret phenology, up to 7 d in var. Moran but in some umbellets of plant F, this phase (Stages h and i) was as short as 2 d (Table 2). Female-phase duration averaged 4.40 ± 0.16 d (N = 40 umbellets) and 6.80 ± 0.43 d (N = 20) for the Karzo and Moran varieties, respectively (P < 0.001).
Post-female phase
The styles and stigmata began to dry up in Stage j, a process completed by Stage k (fig. 11 in Rosengarten, 1969
). Stage j passed within 1 d for certain umbellets, including three in plant F (Table 2). If pollination had occurred during the female phase, then swelling and elongation of ovaries was evident at this time.
Overall, Stage k was reached within as few as 7 d after Stage a (e.g., umbellet F-3 in Table 2), but took 15 d for umbellet B-4 (Table 1). Generally, the duration of flowering was similar between umbellets within a plant (Tables 1 and 2), but differed by up to 4 d (11–15 d in plant B, Table 1). Florets of var. Moran lasted significantly longer (11.50 ± 0.51 d, N = 20 umbellets) than those (8.68 ± 0.18 d, N = 40; P < 0.001) of var. Karzo.
Patterns of daily nectar production
Flanking the base of each style, atop each pale-green inferior ovary, two light-tan floral nectaries were present in each floret (Figs. 1–3). The surface of each nectary contained numerous, uniformly-distributed, sunken stomata (Fig. 1). Except for an increase in epidermal cell size, there was little change in nectary external appearance from Stages b to i. Thereafter, the nectaries gradually turned darker brown.
In most plants, there were two distinct periods of nectar secretion per floret, one in the male phase (Stages b–d) and a second bout in the female phase (Stage h). Nectaries typically remained dry in buds (Stage a) and secretion commenced in Stage b while filaments of stamens elongated, anthers dehisced and petals unfolded. Examination of such florets with the dissecting microscope showed many tiny droplets on the nectary surface emanating from several of the stomatal pores. As these droplets became more voluminous, they coalesced with neighboring ones to yield a surface glistening with exudate by Stage c, as anthers continued to shed pollen.
During the florets' male phase, the majority of umbellets (51 of 60; 85%) yielded their peak nectar quantities in the middle of that phase, in a predictable manner. That is, for most plants, maximal nectar volumes were collected on day 2, except in plants A, B (Table 1), and G (Table 2) where quantities typically peaked on day 3. Seven umbellets (e.g., B-5 in Table 1), which exhibited peak male-phase quantities of nectar on day 1, already had florets predominantly at Stage b, indicating an advancement in flowering compared to that on other umbellets on those plants. Two other anomalies had lowest nectar volumes in the middle of the male phase, thereby displaying a bimodal pattern of secretion.
Concurrent with stamen loss, as florets entered their neutral, intermediate stage (e), and extending through to their stages of style elongation (f, g), nectary surfaces on individual florets were dry for 24 h or more. Absence of any nectar during this period was typical for plant B (Table 1) and plants D, F, and G (Table 2). Indeed, even on an umbellet (ten florets each) basis, 37 of 60 (61.7%) lacked any nectar for at least a full day between male and female phases. Still, asynchrony in floret stages within umbellets resulted in some examples (14 of 60 umbellets; 23.3%) wherein umbellets rated as being between sexual phases yielded collectable nectar: e.g., Stage e (G-2), ef (C-5), f (C-7), fg (C-2, G-2), gf (C-3), and g (B-2) from Tables 1 and 2. One or two lagging, nectar-producing florets at Stage d accounted for the first three instances whereas precocious florets of Stage h (nectar-producing) occurred in the remaining three designations. Therefore, within the same plant, one entire umbellet (B-4) spanned at least two full days without nectar being produced, while a nearby umbellet (B-2) yielded nectar each day between male and female phases (Table 1).
Nectar production during stigmatic receptivity (Stage h) in the female phase was relatively intense initially; peak nectar volumes occurred on the first day of this stage in most (53.3%; 32 of 60) umbellets. This pattern was most pronounced in plants D–H of var. Karzo, where 65% (26 of 40) of umbellets produced their maximal daily volume on the first day of Stage h (Table 2). In fact, in umbellets C-2 (Table 1) and F-3 and F-5 (Table 2), essentially all female-phase nectar was secreted on that first day. At the other extreme, nectar production in the female phase lasted up to 7 d (plant A), sometimes in a bimodal fashion (e.g., umbellet B-4 in Table 1). These longer-duration events resulted from an unusual lack of synchrony among the ten florets per umbellet, their final days with nectar having almost no florets at Stage h; these were rated overall as post-secretory stages (i.e., Stages i, ij, jh, and ji). Otherwise, few umbellets designated as Stages j and k still yielded nectar (Tables 1 and 2).
Umbellets of ten perfect florets produced nectar for as few as 4 d total (F-5 in Table 2), but for as many as 12 d (B-4 in Table 1). On the other hand, nectar production by staminate flowers was low or absent. Only traces of nectar, not enough to measure, were found on staminate florets on any of the plants examined.
Comparative yields of floral nectar among plants
The nectar data from plants of varieties Moran and Karzo are summarized in Tables 3 and 4, respectively. Significant differences in total nectar yield per floret occurred among these randomly chosen plants of each variety. In var. Moran, a 3.1-fold difference occurred between highest and lowest nectar-producers (Table 3); in var. Karzo, plants varied 4.0-fold (Table 4).
|
|
Effects of umbel position and single vs. multiple nectar collections during the female phase
There were significant differences (P < 0.035) between plants in terms of nectar produced in the female phase, nectar yield of plant A exceeding that of plant B in the 3° order (Tables 5 and 6). However, umbel orders at different positions (2° vs. 3° or 4°) on the same plant did not differ significantly (P > 0.15) in nectar yield collected once during their female phases (Table 6), nor did all plants follow the same trend (Table 5).
|
|
The greatest amount of nectar sampled from a single floret at one time was 340 nL during the female phase on a 4° umbel of a growth-chamber plant (var. Moran).
Effect of male-phase nectar removal on female-phase nectar production to follow
Results of nectar collection from five pairs of matched umbellets from each of four plants of var. Karzo are given in Table 7. Interestingly, when nectar was collected daily throughout both male and female phases and compared to florets that were sampled daily only during the female phase, no significant differences in nectar production within the female phase (columns 2 and 5 of Table 7) were detected. No trend was apparent either, nectar values in column 2 being higher than the column-5 values for two plants (F, H), but lower for the others (G, I). However, the overall nectar yield from both sexual phases combined (column 3) was significantly greater than the amount of nectar collected during sampling restricted to the female phase (column 2), as well as for three (F, G, I) of the plants. For plant H, P = 0.131.
|
Nectar-solute concentration and carbohydrate composition
Nectar-solute concentration determined by refractometry was consistently high (range 48–76.5%) and averaged 66.5% (= 66.5 g/100 g) or 84.4 g/100 mL (Búrquez and Corbet, 1991
). There were no significant differences between male and female phases, nor between varieties.
Composition of the three major carbohydrates of caraway nectar are shown in Table 8. Although fructose (F) and glucose (G) were predominant throughout, nectar collected from the male phase of florets differed from that of the female phase in patterns consistent between the two varieties. Whereas the glucose content of nectar stayed relatively constant (24.9–28.7%) among the sexual phases, the levels of fructose and sucrose (S) were greater and lesser, respectively, in female-phase nectar. Accordingly, the F/G ratio was greater, and the S/(F + G) ratio had lessened, during the female phase of nectar production (Table 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Knuth, 1908
; Ricciardelli d'Albore, 1986
), though in spring honey bees favor biennial caraway for its pollen (Langenberger and Davis, 2002
). In our study of annual caraway (var. Moran) in growth chambers without pollinators, no differences occurred in nectar yield per floret with umbel position within the same plants. Under these conditions, seed set in this self-compatible species was low (<10%), in accordance with a previous study (0–12%; Németh, Bernáth, and Petheô, 1999
), and it would be interesting to determine whether or not maternal resources are adequate (Stephenson, 1981
) to allow higher umbel orders to produce as much nectar when lower orders had previously been pollinated and commenced seed filling. There is indirect evidence from indeterminate-flowering, non-apiaceaen species (Andrejeff, 1932
; Fahn, 1949
; Pernal and Currie, 1998
) to support the hypothesis that nectar sugar yields in later flowers are reduced by sparser resources. Similarly, a relationship between low seed set and surplus of photosynthate in nectar had been postulated for Daucus carota (Erickson, Peterson, and Werner, 1979
).
A noteworthy genotypic influence on nectar yield was evident in annual caraway. Within var. Karzo (five plants) and var. Moran (three), 4.0- and 3.1-fold differences, respectively, in average nectar quantity per floret were recorded. Although comparatively sparse quantities (0.51–2.92 µL/10 florets) of nectar were available from the eight randomly chosen plants investigated, the 5.7-fold range in mean nectar yield per floret is not exceptional but reiterates the crop-breeding potential of this trait (Shuel, 1989
; Davis, 2001
). Sampling five umbellets per plant rather than eight was sufficient to determine representative nectar quantity per floret.
Disparate nectar yields during a perfect floret's sexual phases
As well as the differences in nectar yield per floret among plants, clear differences occurred during the two sexual phases within individual florets of caraway. The initial male phase averaged only 26.0% (var. Moran) and 31.0% (var. Karzo), or 29.9% overall, of the total nectar available per floret. This bias to the female phase, in which stigmatic receptivity and nectar secretion lasted for 1–7 d, was modified slightly when nectar yields per phase (grand means of Tables 3 and 4) were expressed on a daily basis. Considering the longer durations of the female phase for which nectar was available (6.8 vs. 3.6 d in var. Moran; 4.4 vs. 3.0 d in var. Karzo), the ratio (female:male) of daily rates of nectar produced per phase are remarkably similar among the two varieties—1.53:1 (0.206:0.135 µL·10 florets–1·d–1) for var. Moran and 1.56:1 (0.178:0.114 µL·10 florets–1·d–1) for var. Karzo. Daily rates of nectar production still favored the female phase.
These nectar data for caraway accord with the few previous studies of nectar production per sexual phase within the Apiaceae. During anther dehiscence in Daucus carota, 0.05 µL nectar per floret was produced over 2 h compared to 0.07 µL at stigmatic receptivity (ratio 1.4:1) (Koul, Sharma, and Koul, 1993
). Moreover, in Coriandrum sativum, Koul, Hamal, and Gupta (1989
, p. 510) stated "the amount of nectar secreted during the receptive phase of the stigma exceeds that secreted during dehiscence of anthers."
In a study of resource allocation to protandrous perfect florets of Smyrnium olusatrum (Apiaceae) investigated until embryos formed, Lovett Doust and Harper (1980)
found an impressive proportion (29.6%) of dry matter (carbon) was allocated to stylopodia (possibly sampled with nectar still intact), followed by stamens (27.9%), petals (24.6%), and pistils (inferior ovaries plus the stigmata and styles excised from the stylopodia; 18.0%) in their untreated plants. While it may only be coincidental, it is nonetheless interesting to note that in S. olusatrum, "The costs of male activities, the biomass of stamens, represent ...1.5 times the investment in pistils at the time of flowering" (Lovett Doust and Harper, 1980
, p. 259). This value opposes the relationship determined above for daily rates of nectar production itself, wherein rates of 1.4-fold in D. carota (Koul, Sharma, and Koul, 1993
) and 1.53–1.56-fold in C. carum (present study) favored female rather than male floral function.
For over 50 yr there has been research devoted to relative nectar production among imperfect flowers of each sex (Fahn, 1949
). In a review of 19 dioecious species, Eckhart (1998)
reported that 10 exhibited higher nectar production in pistillate flowers, the remainder yielding more nectar in staminates. Here, several taxa investigated to examine a relationship between sexual phase of their perfect flowers were surveyed, and a similar split was apparent. Of 19 species (one, Lobelia cardinalis L., was studied twice, independently; see below) exhibiting protandry, 7 offered more nectar during the female phase (Best and Bierzychudek, 1982
; Cruden, Hermann, and Peterson, 1983
; Koul, Sharma, and Koul, 1993
; Gillespie and Henwood, 1994
; present study), 7 produced more nectar during the male phase (Cruden, Hermann, and Peterson, 1983
; Bell et al., 1984
; Devlin and Stephenson, 1985
; Klinkhamer and deJong, 1990
; Snow and Grove, 1995
; Rivera, Galetto, and Bernardello, 1996
; Aizen and Basilio, 1998
), and in 6 there was no significant difference among phases (Cruden, Hermann, and Peterson, 1983
; Pleasants, 1983
). Evidently not recognized previously is the discrepancy posed by flowers of L. cardinalis, a hummingbird-visited species. Studied in Mexico between 1973 and 1975, significantly more nectar sugar was available in the female phase (Cruden, Hermann, and Peterson, 1983
) than in Pennsylvania, northern USA, wherein the male phase produced significantly more nectar sugar for three consecutive years (1980–1982; Devlin and Stephenson, 1985
). Clearly, this species and probably others having populations widespread over expansive geographical ranges provide an important opportunity to determine the plasticity of nectar production with sexual phase in hermaphroditic flowers in the field, an occurrence currently without satisfactory explanation.
Even in the absence of insect visitation and wind, only traces of nectar were encountered on the centrally located staminate florets in annual caraway, thereby being much inferior to the relatively meagre quantities collectable during the male phase of perfect florets. That nectar yields of the latter greatly exceeded those of the later-opening staminate florets may relate to unknown factors favoring femaleness in caraway, as well as competition for resources within the umbellet. Nectary size, which is only slightly reduced in staminate florets of the Apiaceae (Magin, 1983
), is therefore likely an unimportant consideration. Using filter paper discs, Gillespie and Henwood (1994)
succeeded in collecting small quantities of nectar sugar from staminate florets of Polyscias sambucifolia (Araliaceae), which also paled in comparison to those obtained from perfect florets. Of relevance is a recent survey of gynodioecious species in which nectar production by perfect flowers exceeded that of their pistillate counterparts within eight of nine species, the ninth having similar nectar yields in each flower morph (Eckhart, 1998
). Accordingly, nectar production by unisexual flowers is generally lower than that of hermaphroditic flowers.
Grown in controlled conditions, Carum carvi yielded approximately 70% of a perfect floret's total nectar during stigmatic receptivity. Greater nectar production during the female phase in florets of carrot and coriander (Apiaceae) was suggested to compensate, in terms of pollination ecology, for the absence of pollen as a reward (Koul, Sharma, and Koul, 1993
). An additional outcome of superior nectar production in the female phase perhaps is the improved likelihood of insect-borne pollen contacting the wet stigmas that arch above the stylopodia (Fig. 3) when heads and mouthparts of eager visitors are directed to the temporally active, crescent-shaped nectaries below.
It is unknown how insects visiting caraway florets can recognize the exposed nectar. Both visual and odor signals may emanate from the nectar itself or from floral structures. Nectar of Daucus carota fluoresces intensely under UV irradiation and may be distinguished by that property (Thorp et al., 1975
; but see Kevan, 1976
). Vogel (1983)
pointed out that apiaceaen nectar advertises itself by its glistening nature. In the related Polyscias sambucifolia (Araliaceae), Gillespie and Henwood (1994)
postulated that movement of shimmering nectar droplets may be perceived by insect visitors. The gradient in solute concentration occurring within a nectar drop (Heracleum mantegazzianum; Corbet et al., 1979
) may play an important role in these events. Upon contact with the exudate, insect sensory organs permit recognition of the sweet fluid (Vogel, 1983
). Although nectaries of the Apiaceae themselves may exhibit distinctive colors (Behrens, 1879
), unlike caraway the nectaries even may change hue with floret phenology, as in Anethum graveolens (Szujkó-Lacza, 1971
). Such color changes may serve for insects as cues to relative nectar availability. Recently, two species of parasitoid wasps have been shown to respond to odors from nectaries of dill (Patt, Hamilton, and Lashomb, 1999
).
Two periods of nectar production followed by reabsorption and nectar-removal experiments
Isolated aspects of caraway nectar dynamics were identified in other umbelliferous taxa by Behrens and Bonnier in 1879. Behrens found that nectar secretion paused or discontinued, reportedly as a result of wind or rain, but did not attribute nectar disappearance to reabsorption. Bonnier recognized that reabsorption of nectar sugar occurred in Anethum foeniculum, progressively as fruits developed. Besides nectar sugar, reabsorption of both 14C-glutamic acid and 32PO4–3 have been demonstrated in actively secreting nectaries of D. carota (Ziegler and Lüttge, 1959
; Lüttge, 1961
). Persistence of the nectary tissue and maintenance of glandular contact by secreted nectar are prerequisites for nectar reabsorption (Bonnier, 1879
; Búrquez and Corbet, 1991
), and these conditions are met in Carum carvi twice. The same stylopodial glands of perfect florets of caraway underwent secretion, cessation, and reabsorption in each of the two sexual phases separated by an intermediate phase lacking nectar production. The intermediate phase occurred in all perfect florets of caraway, only marginally for some central ones but for up to 5 d in florets forming the periphery of an umbellet. Similarly in Pimpinella anisum, Szujkó-Lacza (1975
, p. 51) stated that the nectaries function for 14 d, "with interruptions." On an umbellet basis in caraway, 62% lacked any nectar for at least a full day, but nectar could be absent from umbellets for much less than 1 d or for over 2 d, depending on the plant. Interestingly, in carrot, entire inbred plants can experience a cessation of nectar production (Erickson and Peterson, 1979
).
Beyond the Apiaceae, interruption of nectar production in an intermediate phase has also been recorded in two species of Araliaceae (Thomson and Barrett, 1981
; Gillespie and Henwood, 1994
). However, unlike two species of Alstroemeriaceae, wherein a flower's nectar production continued for up to a week in a "neuter phase" intervening male and female phases (Snow and Grove, 1995
; Aizen and Basilio, 1998
), caraway florets in the intermediate stage ceased producing nectar and only resumed nectar secretion at stigmatic receptivity. Therefore, if caraway nectar advertises itself (see discussion above), then it may also signal its absence during the floret's intermediate phase.
Regardless of whether the same florets of 2° umbellets were sampled for nectar repeatedly throughout both male and female phases or only had nectar removed for the first time throughout the female phase, no significant difference in female-phase nectar yields occurred in four plants tested. By inference, this outcome establishes that uncollected caraway nectar sugar reabsorbed after completion of the male phase is destined for elsewhere and not to be made available to flower visitors as further nectar secreted at stigmatic receptivity by the same nectaries. Moreover, the more abundant female-phase nectar secretion evidently involves an additional source likely imported to the stylopodia during the stages of late-style elongation and stigma receptivity and forms an independent event from the initial, male-phase bout of secretion from the same glands. A different source or pathway of carbohydrates to the nectaries might also help to explain the change in nectar-carbohydrate composition detected between male and female phases of each variety examined (see below).
Instead of being embellished via full reabsorption of male-phase nectar, nectar yields by caraway umbellets sampled only throughout the female phase were significantly lower than those totals from matched umbellets whose nectar was withdrawn daily throughout both male and female phases. Likewise, nectar quantities from repeated sampling of non-apiaceaen flowers often exceeded nectar yields of flowers sampled less often (see references in Pleasants, 1983
; Búrquez and Corbet, 1991
; Torres and Galetto, 1998
), the differences being attributed to apparent reabsorption (Nicolson, 1995
). Bonnier (1879)
observed in Heracleum sphondylium that floral-nectar production may even continue post-fertilization, in flowers already having lost their petals. He proposed for several species of Apiaceae that nectar constituents, once reabsorbed, are allocated to embryo and fruit development. Interesting in this light are the analyses of N, P, K, and dry matter in Smyrnium florets, wherein "competition within the plant for limiting resources would seem ... to be stronger between pistils and stylopodia than between male and female activities" (Lovett Doust and Harper, 1980
, p. 260). If floral anatomy of several members of the Apioideae studied (Jackson, 1933
) pertains to caraway, then there appear to be vascular connections (traces 3 and 8 in Figs. 6, 13, 14, 17 and 19a; p. 123) that could convey reabsorbed nectar sugar downward, away from the stylopodium and directly to the fertilized ovule, thereby fulfilling the conclusions of Bonnier (1879)
and Lovett Doust and Harper (1980)
. An overlap in niche occupied by the pistils and stylopodia may actually also relate back to the first sexual phase per floret, wherein reclaimed nectar constituents not resecreted during the floret's stage of stigmatic receptivity may instead have contributed to the intervening floral events of style elongation, stigma expansion (cf. Figs. 2 and 3), and stigmatic-exudate production in the same and nearby florets, which is perhaps analogous to the remarkable situation demonstrated with 14C-sucrose in flowers of Streptosolen jamesonii (Solanaceae) discovered by Shuel (1961)
.
Changes in nectar-carbohydrate composition with sexual phase in perfect florets
Floral nectar from both varieties of annual caraway had consistent patterns in composition. Fructose content more than doubled that of glucose, followed by sucrose. Similar fructose-dominant floral nectar in the legume Robinia Pseud-acacia suggested the presence of a transglucosidase in nectar (Zimmermann, 1954
), but it is unknown whether similar enzyme activity occurs in caraway.
Floral nectar composition actually varied with sexual phase of the caraway floret. The mean carbohydrate profile (59.6% fructose, 26.9 % glucose, 13.6% sucrose; S/(F + G) = 0.157) in the male phase is "hexose-rich" but a decline in sucrose with an increase in fructose (70.2% fructose, 26.8% glucose, 3.1% sucrose; S/(F + G) = 0.032) manifests the female-phase nectar "hexose-dominant" (Baker and Baker, 1983
). Whether this change in carbohydrate composition is significant for pollination ecology of caraway is unknown. Both compositions typify those of floral nectars favored by insects like flies, short-tongued bees, and butterflies (Baker and Baker, 1983
), taxa (with wasps) regularly recorded visiting caraway (Knuth, 1908
; Ricciardelli d'Albore, 1986
; Langenberger, 2000
). The predominance of hexoses in nectar, common in shallow, bowl-shaped flowers, causes such exposed nectars to evaporate less readily (Corbet, 1978
), and a thickened layer on apiaceaen nectar droplets assists this resistance (Corbet et al., 1979
).
Floral nectars of the Apiaceae are characterized generally by fructose-glucose dominance (Percival, 1961
). Previous analyses of apiaceaen nectar from the distinct sexual phases of florets, or by HPLC, were not found. Therefore, published results may represent mixtures of nectar from both sexual stages, or a single (unspecified) stage. Nectar-carbohydrate compositions analyzed by paper, thin-layer, or gas-liquid chromatography are recorded for over 20 species of the family. Nectars of Aegopodium podagraria (sFG, Percival, 1961
; sGF, Zauralov and Yakovleva, 1973
; high F and G, low S, Maurizio and Schaper, 1994
), Daucus carota (F/G/S = 52.6/45.7/1.7, Erickson, Peterson, and Werner, 1979
; high F and G, low S, Maurizio and Schaper, 1994
), and Heracleum sphondylium (sFG or SFG, Percival, 1961
; F/G/S = 410/316/56, Finch, 1974
; F/G/S = 47/46/6, Maurizio and Schaper, 1994
) most closely resemble that of caraway. Of these Apioideae, molecular analyses indicate that Aegopodium is most closely related to Carum (Downie et al., 1998
), although even species within a genus can differ widely in nectar-carbohydrate composition (Percival, 1961
). The peak concentrations of two species studied by Finch (1974)
were 945 and 862 µg/µL, similar to those of C. carvi. Based on their high solute concentrations and elevated contents of fructose and glucose with low sucrose levels, Percival (1961
, p. 242) regarded apiaceaen nectar as already "technically a honey."
The intriguing differences in nectar quantity and sugar composition with sexual phase in caraway's perfect florets occur despite involvement of the same stylopodial nectaries. These results may indicate different supplies of carbohydrates entering the nectaries or even variable routes for pre-nectar movement and escape. In our next study the anatomy and ultrastructure of the floral nectaries will be reported from the two nectar-secreting sexual phases, plus the quiescent intermediate stage associated with the marked dichogamy displayed by caraway.
|
FOOTNOTES |
|---|
2 Author for reprint requests (davisa{at}duke.usask.ca
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Andrejeff W. 1932 Über Nektarien und über die Menge des Nektars einiger Gehölzarten. Mitteilungen der Deutschen Dendrologischen Gesellschaft 44: 99-105
Anonymous. 2000 1999 Specialty crops report. Saskatchewan Agriculture and Food, Regina, Saskatchewan, Canada
Baker H. G. I. Baker 1983 A brief historical review of the chemistry of floral nectar. In B. Bentley and T. Elias [eds.], The biology of nectaries, 126–152. Columbia University Press, New York, New York, USA
Behrens W. J. 1879 Die Nectarien der Blüthen: anatomisch-physiologische Untersuchungen. Flora 62: 305-309
Bell G. L. Lefebvre L.-A. Giraldeau D. Weary 1984 Partial preference of insects for the male flowers of an annual herb. Oecologia 64: 287-294[CrossRef][ISI]
Best L. S. P. Bierzychudek 1982 Pollinator foraging on foxglove (Digitalis purpurea): a test of a new model. Evolution 36: 70-79[CrossRef][ISI]
Bieleski R. L. R. J. Redgwell 1980 Sorbitol metabolism in nectaries from flowers of Rosaceae. Australian Journal of Plant Physiology 7: 15-25
Bonnier G. 1879 Les nectaires: étude critique, anatomique et physiologique. Annales des Sciences Naturelles Botanique 8: 5-212
Bouwmeester H. J. H. G. Smid 1995 Seed yield in caraway (Carum carvi). 1. Role of pollination. Journal of Agricultural Science 124: 235-244
Búrquez A. S. A. Corbet 1991 Do flowers reabsorb nectar?. Functional Ecology 5: 369-379
Corbet S. A. 1978 Bees and the nectar of Echium vulgare. In A. J. Richards [ed.], The pollination of flowers by insects, 21–29. Academic Press, London, UK
Corbet S. A. P. G. Willmer J. W. L. Beament D. M. Unwin O. E. Prys-Jones 1979 Post-secretory determinants of sugar concentration in nectar. Plant, Cell and Environment 2: 293-308
Cruden R. W. S. M. Hermann S. Peterson 1983 Patterns of nectar production and plant-pollinator coevolution. In B. Bentley and T. Elias [eds.], The biology of nectaries, 80–125. Columbia University Press, New York, New York, USA
Davis A. R. 1997 Influence of floral visitation on nectar-sugar composition and nectary surface changes in Eucalyptus. Apidologie 28: 27-42
Davis A. R. 2001 Searching and breeding for structural features of flowers correlated with high nectar-carbohydrate production. Acta Horticulturae 561: 107-121
Davis A. R. V. K. Sawhney L. C. Fowke N. H. Low 1994 Floral nectar secretion and ploidy in Brassica rapa and B. napus (Brassicaceae). I. Nectary size and nectar carbohydrate production and composition. Apidologie 25: 602-614[CrossRef][ISI]
Devlin B. A. G. Stephenson 1985 Sex differential floral longevity, nectar secretion, and pollinator foraging in a protandrous species. American Journal of Botany 72: 303-310[CrossRef][ISI]
Downie S. R. S. Ramanath D. S. Katz-Downie E. Llanas 1998 Molecular systematics of Apiaceae subfamily Apioideae: phylogenetic analyses of nuclear ribosomal DNA internal transcribed spacer and plastid rpoC1 intron sequences. American Journal of Botany 85: 563-591[Abstract]
Eckhart V. M. 1998 Sexual dimorphism in flowers and inflorescences. In M. A. Geber, T. E. Dawson, and L. F. Delph [eds.], Gender and sexual dimorphism in flowering plants, 123–148. Springer-Verlag, Berlin, Germany
Erickson E. H. C. E. Peterson 1979 Asynchrony of floral events and other differences in pollinator foraging stimuli between fertile and male-sterile carrot inbreds. Journal of the American Society for Horticultural Science 104: 639-643
Erickson E. H. C. E. Peterson P. Werner 1979 Honey bee foraging and resultant seed set among male-fertile and cytoplasmically male-sterile carrot inbreds and hybrid seed parents. Journal of the American Society for Horticultural Science 104: 635-638
Fahn A. 1949 Studies in the ecology of nectar secretion. Palestine Journal of Botany, Jerusalem Series 4: 207-224
Finch S. 1974 Sugars available from flowers visited by the adult cabbage root fly, Erioischia brassicae (Bch.) (Diptera, Anthomyiidae). Bulletin of Entomological Research 64: 257-263
Gillespie L. H. M. J. Henwood 1994 Temporal changes of floral nectar-sugar composition in Polyscias sambucifolia (Sieb. Ex. DC) Harms (Araliacaeae). Annals of Botany 74: 227-231
Harder L. D. S. C. H. Barrett 1992 The energy cost of bee pollination for Pontederia cordata (Pontederiaceae). Functional Ecology 6: 226-233
Jablonski B. Z. Koltowski 1993 Nectar secretion and honey efficiency of important honey plants growing under Poland's conditions. Pszczelnicze Zeszyty Naukowe 37: 81-87
Jackson G. 1933 A study of the carpophore of the Umbelliferae. American Journal of Botany 20: 121-144[CrossRef][ISI]
Kevan P. G. 1976 Fluorescent nectar. Science 194: 341-342
Klinkhamer P. G. L. T. J. de Jong 1990 Effects of plant size, plant density and sex differential nectar reward on pollinator visitation in the protandrous Echium vulgare (Boraginaceae). Oikos 57: 399-405[CrossRef][ISI]
Knuth P. 1908 Handbook of flower pollination. Clarendon, Oxford, UK
Koul A. K. I. A. Hamal S. K. Gupta 1989 Pollination mechanism in Coriandrum sativum Linn. (Apiaceae). Proceedings of the Indian Academy of Sciences (Plant Sciences) 99: 509-515
Koul P. N. Sharma A. K. Koul 1993 Pollination biology of the Apiaceae. Current Science 65: 219-222
Langenberger M. W. 2000 Studies of caraway pollination by honey bees and other insects in Saskatchewan. Master's thesis, University of Saskatchewan, Saskatoon, Saskatchewan, Canada
