Capsule Dehiscence in Viola betonicifolia Sm. (Violaceae) R. John Little1 & Glenn Leiper2 Summary Little, J. & Leiper, G. (2012). Capsule dehiscence in Viola betonicifolia Sm. (Violaceae). Austrobaileya 8(4): 624-633. Seed dispersal syndromes in Viola are reviewed and the sequence of events culminating in the dispersal of seeds from capsules of Viola betonicifolia is documented. Seed parameters (length x width) and measurements of distances travelled after being ballistically ejected from a capsule valve were determined. Preliminary observations were recorded of the approximate length of time for a mature capsule to open and the approximate time for an open capsule to eject all seeds from its three valves. Key Words: Violaceae, Viola, Viola betonicifolia, capsule dehiscence, diplochory, myrmecochory 'R John Little, MagnaFlora, 16 Pebble River Circle, Sacramento, California 95831, U.S.A. Email: [email protected] 2Glenn Leiper, Tweedvale Street, Beenleigh, Queensland 4207, Australia. Email: leiper30@,bigpond. com Introduction The effective dispersal of seeds to sites with walls thickened with sclereids that where seedlings can successfully establish is produce pressure on seeds; seeds with a small critical to most vascular plants that reproduce elaiosome; and seeds that are ballistically sexually (Forster 2007). Understanding dispersed 1-5 m from the parent plant. A the methods by which seeds are dispersed third type, called ballistic or autochory provides valuable information about a species’ (explosive ejection of seeds away from the potential for distribution and colonisation, parent plant not related to ant dispersal) has and potential taxonomic relationships. Two been suggested for Viola, but was not detected distinct seed dispersal syndromes have been as a distinct syndrome in the species studied recognised in Viola. 1. myrmecochory (seed by Beattie & Lyons (1975). dispersal by ants following passive release Perhaps the first documented observation from capsules) and 2. diplochory (explosive of capsule dehiscence in Viola was that of ejection of seeds followed by attraction and Leavitt (1902) for Viola rotundifolia Michx., a dispersal by ants) (Beattie & Lyons 1975). perennial species from North America. Based Characters attributable to the myrmecochory on the distance that seedlings germinated in syndrome include a prostrate peduncle during his garden from a single ‘mother’ plant, the dehiscence with the capsule on or near the species appeared able to ballistically eject its ground; a calyx often swollen during fruiting; seeds 1.5-2.7 m (Leavitt 1902). Harrington a capsule with walls not greatly thickened; (1903) stated, “after dehiscence [the capsule seeds never under pressure during dehiscence; valves] fold lengthwise and eject the seeds seeds with a large and conspicuous elaiosome; with some force,” and Bare (1979) said, “As and seeds that are dispersed passively beneath each segment dries, it slowly folds together the parent plant. Characters attributable to the lengthwise, and the resultant pressure on the diplochory syndrome include a tall peduncle seeds throws them several feet through the that is erect during dehiscence and carries the air.” capsule above the leaves; a calyx that does not enlarge during fruit development; a capsule The method by which the capsules of diplochorous Viola species dehisce has been described variously as “sometimes Accepted for publication 9 November 2012 dehiscing explosively” (Lawrence 1951), Little & Lei per, Capsule dehiscence in Viola betonicifolia 625 “the matured capsule opens and ejects the Questions we attempted to answer in this seeds in a moment” (Ohkawara & Higashi study for Viola betonicifolia included: 1994), “seeds are explosively ejected from the • What is the sequence of events for the capsules” (Douglas & Ryan 1998), “opening peduncle and capsule leading up to seed ± explosively on maturity” (Thiele & Prober dispersal? 2003), and “at maturity the capsule springs • Does the inverted peduncle of open and the seeds are forcefully ejected” cleistogamous capsules remain pendant (Karlsson etal. 2012). These descriptions give or become erect? the impression capsules ‘explode’ to release • How long does it take for a mature their seeds, in a manner similar to some genera capsule to open? in the Euphorbieae (Forster 2007). These • How long before seeds begin to disperse descriptions obscure the actual mechanism from an open capsule? by which capsules of many Viola species open • Are seeds dispersed randomly from and how the seeds are subsequently dispersed. a valve, or is there a pattern (e.g., do Other workers describe how seeds are seeds disperse first from the center, the released from the capsule, e.g., “explosively proximal, or the distal ends of a valve)? ejected”, “ballistically dispersed”, or • What are minimum and maximum “ballistically ejected”, which more accurately dispersal distances? describes how seeds are dispersed (Turnbull & Culver 1983; Bulow-Olsen 1984; Brooks & For comparison with other Viola species McGregor 1986; Douglas & Ryan 1998; Little we documented the number of seeds produced & McKinney in prep). However, there appear by cleistogamous flowers (abbreviated CL, to be no detailed studies for diplochorous which are self-pollinated flowers, with minute species documenting the actual process by or no petals); measurements of seed lengths; which Viola capsules open and how the seeds and morphological details of chasmogamous are dispersed. flowers (abbreviated CH, which are flowers with showy petals). This note focuses on Viola betonicifolia Sm. subsp. betonicifolia, a species native to Materials and methods Australia. This species possesses characters There were two rounds of observations. attributable to the diplochorous seed dispersal The first author observed and photographed syndrome, i.e., a tall peduncle with the capsule Viola betonicifolia plants near Mt Lindesay above the leaves at the time of dehiscence, a in south-eastern Queensland, in December calyx that does not enlarge when the capsule 2011 and 20 January 2012 during visits with develops, seeds with a small elaiosome, the second author. Initial observations of and seeds that are ballistically ejected some capsules opening and seeds being ballistically distance from the parent plant. ejected were made by the first author on 22 During a visit by the first author to January 2012 from a plant collected as a Queensland and New South Wales, Australia, voucher specimen on 20 January 2012, but from November 2011 through January 2012, an not yet pressed. In the field, the collected opportunity arose to observe mature capsules plant was placed in a cup of water. The plant of Viola betonicifolia and V. he derace a Labill. had two capsules which for convenience are splitting open and to observe how seeds were discussed herein as A and B. Capsule A was ejected from capsules of V betonicifolia. The photographed at different intervals after it second author made numerous additional had opened until all seeds were ejected. The observations in 2012 of how capsules open number of seeds in each valve was determined in V betonicifolia and documented distances for Capsule A, as was the approximate length that seeds were ejected. of time from when it began to open until all seeds were ejected. While attention was focused on Capsule A, Capsule B had opened and ejected an unknown number of seeds. 626 Austrobaileya 8(4): 624-633 (2012) Observation of Capsule B began when it and Tasmania to Queensland; it is also found had a total of 10 seeds remaining among the in New Guinea and the Philippines. It occurs three valves. For both capsules, the time of in a wide range of habitats from coastal dunes day and number of seeds remaining in each and sclerophyllous forest to alpine herbfields valve were recorded at periodic intervals and (Adams 1982). The species produces both documented with photographs. CH and CL flowers. The species is available for horticulture commercially and is sold in Although valuable data were obtained Australia mainly through specialist native during this first period of observation by the plant nurseries. first author, dispersal distances could only be roughly approximated due the fact that it Floral Morphology was difficult to observe where the dehisced Fig. 1 is a typical CH flower of Viola seeds had landed. Dr Paul Forster encouraged betonicifolia. The lateral petals are bearded the authors to gather better data on dispersal (Adams 1982; James 1990a,b). We found that distances. The second author volunteered CH flowers on some plants at Mt Lindesay to continue a second round of observations also have hairs on the two upper petals (Fig. and to measure dispersal distances and 1), which has not previously been reported. seed parameters. Eventually, container- The CL flowers of V betonicifolia have sepals grown plants were obtained and maintained but no petals (Fig. 2). outdoors by the second author in Beenleigh, Queensland. During this study, no CH flowers were present. Plants with CL capsules that appeared ready to open were brought indoors and dispersal distances were able to be measured under controlled circumstances. To determine dispersal distances, white flannelette sheets with roughened surfaces to minimize bounce of seeds were placed around the containers and the distances where seeds landed were measured by two observers. Photographs were taken of maturing capsules to document a) the movement of the peduncle; Fig. 1. Typical CH Viola betonicifolia flower; Mt b) the position of capsules prior to splitting Lindesay, Queensland. Photo: J. Little and opening; c) whether capsules open slowly or explosively; d) the number of seeds per capsule; and e) if seeds are randomly dispersed from a valve or if there is a pattern by which they are dispersed. Seed measurements (length x width) were made by the second author with a micrometer. Statistics were calculated with Microsoft® Excel. Results Fig. 2. A CL flower developing from the base of the Viola betonicifolia subsp. betonicifolia is a plant. Sepals are visible enclosing the ovary. Petals are perennial, acaulescent species native to the absent. Photo: G. Leiper east coast of Australia from South Australia Little & Leiper, Capsule dehiscence in Viola betonicifolia 627 Capsules After the ovules are fertilized, the capsules of CL flowers begin maturing in an inverted position, typical of CL flowers of many Viola species. As the capsule matures the peduncle elongates, lifting the capsule to an elevated position above many of the leaves. It transitions from an inverted to an upright position in about 24 hours (Fig. 3). The capsule also becomes noticeably paler, possibly because the xylem ceases functioning. The capsule becomes dehydrated to facilitate splitting open and eventual seed dispersal. The number of fertile seeds can be counted when the capsule is upright. In about another 24 hours, the capsule begins to split open (Fig. 4), the three valves begin to spread back and become completely separated (Fig. 5). Eventually the valves become more or less parallel to the ground (Fig. 6). Occasionally, a few seeds are ejected before the capsule is fully open. Fig. 4. Capsule in fully upright position beginning to split open. Note shrivelled sepals. Photo: G. Leiper Fig. 3. Capsule from CL flower on peduncle transitioning from inverted to upright position. Photo: G. Leiper As the capsule valves begin to dry, the edges move toward each other. The pressure of the constricting valves squeezes the seeds which eventually causes them to be ballistically ejected. We observed that seeds at the distal end of valves were ejected first (Fig. Fig. 5. Capsule completely split open with valves 6) and those at the proximal end were ejected beginning to spread apart. Each valve is ca. 13 mm long. last (Fig. 7). However, for other capsules Photo: J. Little 628 Austrobaileya 8(4): 624-633 (2012) Fig. 6. Seeds in a valve about midway through dehiscence. Note the distal end of these valves contracted before the proximal end. Photo: J. Little Fig. 7. One valve completely dehisced; 2 valves with one seed remaining at proximal end of valve. Photo: J. Little seeds at the proximal end were ejected first. Most capsules observed in this study (n=7) ejected all their seeds (Fig. 8). Observations of Seed Dehiscence Capsule A (recorded by first author): Capsule A was first observed open and in an upright position at 1121 with a total of 35 seeds (Table 1) . The length of time it took to transition from an inverted to upright position was not observed nor was the time noted when the first seed was ejected. All seeds were ejected from the capsule in about 2.3 hr (138 minutes). Valve 1, which initially had two more seeds than the other valves, retained its seeds for a longer period of time. Valve 3 was the first to eject all its seeds. Capsule B (recorded by first author): This Fig. 8. Valves fully dehisced and completely clasped shut. capsule was first noticed when it was open Photo: J. Little and after it had ejected an unknown number of seeds. Observations of Capsule B began at The second author timed a capsule that 1603 when it had 10 seeds remaining (Table started opening at around 1400, was almost 2) . Nine seeds were dispersed in the next 23 fully open at 1445, and was fully open at minutes. The time when the last seed was 1530. During the 45 minute interval between ejected from Valve 3 was not observed, but no 1445 and 1530, some seeds were observed seeds were present at 1715. being ejected from the capsule. This capsule took about 1.5 hr to fully open and all seeds had been ejected by 1610. Thus, after it fully Little & Lei per, Capsule dehiscence in Viola betonicifolia 629 Table 1. Seed dispersal times of Viola betonicifolia (Capsule A) No. of Seeds in Valve Time Valve 1 Valve 2 Valve 3 Total 1121 13 11 11 35 1300 9 4 6 19 1303 8 4 6 18 1304 8 4 5 17 1311 5 4 0 9 1312 5 0 0 5 1338 2 0 0 2 1339 0 0 0 0 Table 2. Seed dispersal times of Viola betonicifolia (Capsule B) No. of Seeds in Valve Time Valve 1 Valve 2 Valve 3 Total 1603 4 4 2 10 1604 4 3 1 8 1610 0 1 1 2 1626 0 0 1 1 1715 0 0 0 0 opened, it took 40 minutes until all seeds were ejected (some seeds dispersed before it was fully open). Seeds In the ovary, the ovoid-shaped seeds are attached to the placenta at the narrow end of the seed at the location of the elaiosome. Typical of many Viola species, the seeds are smooth and shiny when fully mature and fresh (Fig. 9). Mature seeds appear black or brown. Under magnification, they appear mottled black and brown (Fig. 9). The elaiosome is white or whitish when fresh (Fig. 9). The Fig. 9. Viola betonicifolia seeds emphasising elaiosomes. Lines are mm. Photo: J. Little mean number of seeds per CL capsule was 19 (n=7 capsules, 133 seeds; range 8-35; SD crushed when measured. The mean seed 8.42; Table 3). The mean number of seeds per width was 1.11 mm (n=74 seeds, five capsules); capsule is considered preliminary; additional range 0.98-1.30 mm; SD 0.075. The width of counts need to be made utilizing a larger data six seeds was not included because the colour set. Raw data for seed lengths and widths are of five was markedly paler than other seeds available from the authors. in the capsule and thus were assumed to not Mean seed length was 1.44 mm (n=75 be fully mature and/or possibly not viable; seeds, five capsules); range 1.10-1.71 mm; another seed was inadvertently crushed when SD 0.160. The length of five seeds was not measured. determined because they were inadvertently 630 Austrobaileya 8(4): 624-633 (2012) Dispersal Distances Table 3. Numbers of seeds per capsule in Viola betonicifolia The distances that seeds travelled after being ballistically ejected from capsule valves are Capsule No. Number of seeds summarized in Table 4. The mean distance for all seeds (n=85) was 148 cm; range 13-321 1 15 cm; SD 77.76. 2 15 3 19 The majority of seeds were dispersed a 4 8 distance of 101 to 200 cm (Fig. 10). Two seeds 5 23 exceeded 300 cm (306 and 321 cm). 6 18 7 35a a Determined by first author; all others by second author. Table 4. Seed dispersal distances in Viola betonicifolia Dispersal Temp Total number Mean distance/ Distances (cm) Date (°C) of seeds per Range/ Standard capsule Deviation (SD) 19 February n/a 11 78.04 cmJ 17, 33.5, 37.5, 50, 72, 76.5, 95, 98.5, 2012 17-154.5 cm; SD 106, 118, 154.5 41.36 22 February 28.5 19 140 cm/13-256 13, 13.5, 56, 85, 90.5, 105, 132.5, 2012 cm; SD 66.36 134.5, 139.5, 157, 159, 168, 174.5, 175, 175.5, 188.5,210, 226.5, 256 1 March 31.5 24 158.02 cm/ 27- 27, 52.5, 55.5, 62, 79, 93.5, 104, 128, 2012 253 cm; SD 67.85 133.5, 146.5, 161, 180, 188, 191.5, 198.200.5, 204,211.5,213,213, 224.5, 226, 247, 253 2 March 31 15 181.16 cm/22.5- 22.5, 83, 103, 141.5, 144, 150, 166.5, 2012 321 cm; SD 78.62 177, 216, 217.5, 218, 226, 236.5, 295, 321 8 March 28 16 161.06 cmJ 20- 20, 33, 49.5, 76, 96.5, 106.5, 121.5, 2012 306 cm; SD 97.45 141, 152.5, 177.5, 208.5,215, 284, 294, 295.5, 306 Discussion known to produce CL flowers, nine species do not, and the condition in four species remains Cleistogamous flowers (CL) unknown (Little & McKinney in prep). The We documented the presence of CL flowers in number of Viola species in the Australian Viola betonicifolia. The presence or absence of flora that produce CL flowers has not been CL flowers in this species is not mentioned in determined. most Australian floras or guidebooks (Adams The fact that all CL capsules examined 1982; Stanley & Ross 1983; James 1990a,b; in this study contained mostly fertile seeds Robinson 1994; Entwisle 1996; Fairley & suggests that the pollination mechanism in Moore 2000; Anon 2007; Duretto 2009; CL flowers of Viola betonicifolia is highly Elliot & Jones 2010). While not specifically efficient. Mayers & Lord (1983a,b) reported an mentioning cleistogamous flowers in V interesting situation for V. odorata L. where betonicifolia, Williams (1979) referenced the the pollen grains in CL flowers germinate presence of “self-pollinating flowers”. Of the while still in the undehisced anther sacs; the 73 Viola species in North American, 60 are Little & Lei per, Capsule dehiscence in Viola betonicifolia 631 w -D <U cu to I E 3 1-25 26-100 101-200 201-300 300+ Distances (cm) Fig. 10. Dispersal distance frequency histogram (n=84) for Viola betonicifolia pollen tubes then penetrate the sac and grow of diplochorous species that eject seeds out towards the nearby stigma. Karlsson et al. ballistically. Upon drying, contraction of (2012) stated that preliminary observations each of the three capsule valves squeezes the in other Viola species suggest that pollen is seeds, which are then ballistically ejected. simply released in close proximity to the We observed that V betonicifolia seeds were stigma on the recurved style. The mechanism sometimes ejected before the valves were by which CL flowers of V betonicifolia achieve completely open. This occurrence needs to be self-pollination is unknown and needs study. taken into account when determining the total number of seeds per valve and per capsule. Capsules The morphology of mature capsules of Observations of mature capsules of Viola CH and CL flowers are indistinguishable. betonicifolia and other Australian species by Only CL capsules were available for this the authors, and unpublished observations study. Dispersal in CH capsules has not been by the first author on several Californian studied. However, based on observations species, show that the capsules of putatively of CH capsules of other Viola species in diplochorous Viola species do not open Australia and California, we suspect that ‘explosively,’ but instead open rather slowly. CH capsules of V betonicifolia dehisce in a Based on this study and observations of other manner similar to CL capsules. Although Viola species, the capsule itself does not empirical data are not available, it seems ‘explode.’ plausible that temperature, humidity, and soil Capsules of Viola species that disperse moisture are important factors in the length of seeds ballistically are usually on erect time it takes for a capsule to change from an peduncles, whereas capsules that passively inverted to an upright position, to split open, release their seeds (e.g. V odorata) usually and to begin dehiscing seeds. point downward (Beattie & Lyons 1975). How long does it take a mature capsule to Capsules of V betonicifolia are on erect open? Based on one capsule that was timed, it peduncles prior to seed dispersal, typical took 1.5 hr to fully open. Although only one 632 Austrobaileya 8(4): 624-633 (2012) capsule was timed, it was observed that all & Lyons’ (1975) data with V betonicifolia was capsules opened slowly. None ‘exploded’ in a made by considering only the results from CL spontaneous release of seeds. After a capsule capsules: V striata, mean dispersal distance is fully open, how long before seeds begin to was 110 cm (n=50), range 20-220 cm; V disperse? Data from two capsules showed that blanda, mean dispersal distance was 80 cm one took 2.3 hr while another took 40 minutes (n=18), range 30-220 cm; and Vpapilionacea, to disperse all their seeds. Are seeds randomly mean dispersal distance was 100 cm (n=527), dispersed from a valve or is there a pattern? range 2-210 cm. The mean dispersal distance We observed that seeds were dispersed first of V betonicifolia seeds from CL capsules from the centre and then randomly from (149 cm) exceeded the three species reported either the proximal or distal ends of the valve. by Beattie & Lyons (1975) by 110, 80, and Beattie & Lyons (1975) observed that seeds 100 cm. In addition, the maximum dispersal of Viola species they studied were usually distance of V betonicifolia (321 cm) exceeded dispersed first from the centre of the valve. the three species reported by Beattie & Lyons (1975) of 220, 220, and 210 cm. Seeds and Dispersal Distances Forster (2007) documented that The position of the individual valves relative ballistically ejected seeds of Euphorbia to ground level prior to dehiscence affects the obesa Hook. (Euphorbiaceae) were viscid trajectory and dispersal distance for a given resulting in their sticking to soil, pebbles seed. The seed dispersal distances that we and vegetation. We found Viola betonicifolia report are measures of the horizontal distance seeds were buoyant in water, but not viscid from the plant and may not reflect the actual and thus do not stick to surfaces. In coastal distance travelled. For example, a seed could areas of Queensland, V betonicifolia grows be ejected 200 cm vertically, but land only 20 with Melaleuca quinquenervia (Cav.) cm from the plant. The dispersal distance in S.T.Blake in swampy areas that become this case would be measured as 20 cm. We seasonally or periodically inundated from did in fact observe that some seeds were rainwater. In addition to ants, seed dispersal expelled vertically while others were expelled may be facilitated by being buoyant and ± horizontally. Further research is needed to transported in water during storm events. The determine if position in the valve correlates potential that dispersal of Viola seeds can be with the distance a seed is ejected, other facilitated by water has not been investigated. factors being equal. Anecdotal observations of V betonicifolia The mean dispersal distance for all Viola plants growing in Tines’ in coastal wetlands betonicifolia seeds was 149 cm (n=84), range in Queensland where water has obviously 13-321 cm. The frequency histogram of carried them and receded (pers. comm., F. dispersal distances approximates a bell curve Jordan), suggests that water dispersal may (Fig. 10). Dispersal distances of seeds from be an important, heretofore overlooked V betonicifolia CH flowers remain to be mechanism for dispersal of this species in determined. Australia. Beattie & Lyons (1975) reported distances Acknowledgements of ballistically dispersed seeds for CH The authors appreciated the comments of K.R. capsules of seven perennial Viola species Thiele on an earlier draft and encouragement native to eastern North America. They also of P.I. Forster to conduct additional research. reported data for CL capsules for three of the seven species they studied (V striata Aiton, V blanda Willd. and V papilionacea = V. sororia Willd. var. sororia). Because dispersal distances of seeds from CH capsules were longer for each species compared to CL capsules, a reasonable comparison of Beattie Little & Lei per, Capsule dehiscence in Viola betonicifolia 633 References -(1990b). Violaceae. In G.J.Harden (ed.). Flora of New South Wales 1: 435-441. Royal Botanic Adams, L.G. (1982). Viola. In A.S.George (ed.). Flora Gardens: Sydney. of Australia 8: 91-100. Australian Government Publishing Service: Canberra. Karlsson, T., Marcussen, T., Wind, P. & Jonsell, B. (2012). Violaceae. In Nordic Flora, version Anon. (2007). Electronic Flora of South Australia 6b 20080627. http://www.floranordica.org/ species Fact Sheet, http://www.flora.sa.gov.au/ Review/-Review_public/accounts/Viola-x.html cgi-bin/speciesfacts_display.cgi?form=species Accessed 1 Jun 2012. facts&family=&genus=Viola&species=betoni cifolia&iname=&submit=Display Accessed 1 Lawrence, G.H.M. (1951). Taxonomy of Flowering Aug 2012. Plants. Macmillan: New York. Bare, J.E. (1979). Wildflowers and Weeds of Kansas. The Leavitt, R.G. (1902). Seed dispersal of Viola rotundifolia. Regents Press of Kansas: Lawrence, Kansas. RhodoraA. 183-184. Beattie, A.J. & Lyons, N. (1975). Seed dispersal in Little, R.J. & McKinney, L.E. In prep. Violaceae. Viola (Violaceae): adaptations and strategies. Flora of North America Editorial Committee American Journal of Botany 62: 714-722. (eds.). Flora of North America North of Mexico, Volume 6. Flora of North America Association: Brooks, R.E. & McGregor, R.L. (1986). Violaceae. In New York and Oxford. Flora of the Great Plains. The Great Plains Flora Association: Lawrence, Kansas. Mayers, A. & Lord, E. (1983a). Comparative flower development in the cleistogamous species Viola Bulow-Olsen, A. (1984). Diplochory in Viola: A odorata. I. A growth rate study. American possible relation between seed dispersal and Journal of Botany 70: 1548-1555. soil seed bank. American Midland Naturalist 112: 251-260. -(1983b). Comparative flower development in the cleistogamous species Viola odorata. II. An Douglas, G.W. & Ryan, M. (1998). Status of the organographic study. American Journal of yellow montane violet, Viola praemorsa ssp. Botany 70: 1556-1563. praemorsa (Violaceae) in Canada. Canadian Field-Naturalist 112: 491-495. Ohkawara, K. & Higashi, S. (1994). Relative importance of ballistic and ant dispersal in two diplochorous Duretto, M.F. (2009). Violaceae. In M.F.Duretto (ed.). Viola species (Violaceae). Oecologia 100: 135- Flora of Tasmania Online, http://demol.tmag. MO. tas.gov.au/treatments/families/Violaceae/ Violaceae.html Accessed 1 Aug 2012. Robinson, L. (1994). Field Guide to the Native Plants of Sydney (Revised 2nd Edition), p. 224-225. Elliot, R. & Jones, D.L. (2010). Encyclopedia of Kangaroo Press: Kenthurst. Australian plants suitable for cultivation, 9: 440-444. Lothian Books: Melbourne. Stanley, T.D. & Ross, E.M. (1983). Flora of south¬ eastern Queensland 2: 99-101. Queensland Entwisle, T. J. (1996). Violaceae (except Hymenanthera). Department of Primary Industries: Brisbane. In N.G.Walsh & T.J.Entwisle (eds.). Flora of Victoria 3: 361-367. Inkata Press: Melbourne. Thiele, K.R. & Prober, S.M. (2003). New species and a new hybrid in the Viola hederacea species Fairley, A. & Moore, P. (2000). Native Plants of the complex, with notes on Viola hederacea Labill. Sydney District. An Identification Guide Muelleria 18: 7-26. (Revised Edition), p. 84-85. Kangaroo Press: Kenthurst. Turnbull, C.L. & Culver, D.C. (1983). The timing of seed dispersal in Viola nuttallii: Attraction Forster, P.I. (2007). Seed dispersal in the Euphorbieae of dispersers and avoidance of predators. (Euphorbiaceae). Euphorbia World 3: 26-30. Oecologia 59: 360-365. Harrington, H.D. (1903 reprinted 1979). Manual of the Williams, K. (1979). Native Plants of Queensland, Plants of Colorado. Reprinted for Grove Press Volume 1. K.A.W. Williams: North Ipswich. by University Microfilms International: Ann Arbor, Michigan. James, T.A. (1990a). Viola betonicifolia Sm. In New South Wales Flora online. http://plantnet. rbgsyd.nsw.gov.au/cgi-bin/NSWfl.pl7page =nswfl&lvl=sp&name=Viola~betonicifolia Accessed 1 Oct 2012.