Abstract
Passiflora incarnata L., commonly known as maypop, is a wild passion fruit native to many areas of the eastern and southern United States where the climate ranges from subtropical to temperate. Although P. incarnata has had little attention paid to it for breeding purposes, it could be used in breeding for fruit production and possibly contribute cold hardiness genes in combination with other Passiflora species. The study was performed in 2018, 2019, and 2021 at the Mississippi State University South Mississippi Branch Experiment Station in Poplarville, MS, United States. Passiflora propagules were collected from various locations: Florida (FL), Illinois (IL), Mississippi (MS), Missouri (MO), and Oklahoma (OK). Of the 122 flowers across the five P. incarnata genotypes from differing locations, none of them produced a fruit or had any indication of successful or partially successful fertilization when selfed, indicating strong self- incompatibility. If self-compatibility does exist in nature, it is likely to be rare. However, certain combinations of P. incarnata from different locations produced successful fruiting, including IL × MO (52% success), FL × MO (85%), FL × OK (80%), MS × OK (40%), MO × IL (50%), MO × OK (40%), and OK × MO (80%). The differences across provenances show that incompatibility exists within P. incarnata but can depend on location. Overall, fruit weight, fruit size, and soluble solids content measured in this study were similar to and, in some cases, greater than those previously reported. These differences help to illustrate the diversity within P. incarnata and the still-untapped potential for breeding improvements. The problem of self-incompatibility is complex and there is much to learn about how Passiflora species, especially P. incarnata, function. Much of the U.S. domestic market is not familiar with passion fruit, especially as a table fresh product. This could be a barrier to adoption, but it could also prove to be an opportunity to create a niche within the present market and expand it. Although maypop fruit quality is not equal to that of Passiflora edulis Sims, selecting superior wild genotypes with desirable attributes to be used in future intra- and interspecific breeding is possible based on the results of this study.
Passiflora is a large genus that comprises numerous species. It is primarily a tropical crop, grown in regions with mild winter temperatures; however, there are temperate and subtropical Passiflora species, most of which are exclusively known for their unique flowers. One prominent exception is Passiflora incarnata, commonly known as maypop, a wild passion fruit native to many areas of the eastern and southern United States (Killip, 1938). Historical documentation of Passiflora species in North America is limited (Hermsen, 2021); however, archaeological evidence has placed P. incarnata as a prominent food crop for Native American populations since the Late Archaic period (Gremillion, 1989; Hoch, 1934), thus identifying it as an important dietary supplement for indigenous peoples. These groups may have attempted to domesticate it as a stable food crop as well (Asch and Hart, 2004; Gremillion, 1989). Although not broadly grown as a fruit crop, European settlers also consumed maypops. Present day consumption of maypop is limited to local foragers and hobbyists who prize the fruit for its unique flavor. Yet, with the current increase in the Hispanic population throughout the United States, there is a base of people who are familiar with the flavors that passion fruit can offer because many Passiflora species are native to Latin America (Jorgensen et al., 1984). This segment of the U.S. population could provide a built-in market for new and improved passion fruit cultivars.
In addition to its potential nutritive properties, P. incarnata has been used extensively for medicinal purposes. In many areas of the world, it is used as a sedative and anxiolytic (Dhawan et al., 2001) along with other uses (Dhawan et al., 2004). Although broadly used for herbal medicine, toxicity to extracts of P. incarnata has been noted (Fisher et al., 2000). Thus, the ramifications of consuming maypop as a food product merit further study.
Although the vine produces exquisite flowers and an edible fruit, it is primarily considered a weed that can interfere with production of crops such as peanuts (Arachis hypogaea L.) (Wehtje et al., 1985). The vine has a perennial root system and is often found in disturbed sites because it resprouts readily from cut root pieces (Tague and Foré, 2005; Wehtje et al., 1985). The aboveground portion of the vine dies back to the soil line in autumn because of various elements, such as exposure to cold temperatures, reduction in daylength, and disease or insect pests, although all potential causes have not been fully explored. Flowers of this species are mostly hermaphroditic, herkogamic, and functionally andromonoecious (Dai and Galloway, 2011, 2012; Krosnick et al., 2017; Spears and May, 1988). These conditions contribute to self-incompatibility, thus requiring pollination from insects, primarily large-bodied bee species (Krosnick et al., 2017; McGuire, 1999).
Fruit-bearing Passiflora cultivars have been developed mainly from Passiflora edulis f. edulis and P. edulis f. flavicarpa; however, very little crop production is done in more temperate climate zones because of the lack of cold hardiness in these species (Knight, 1971). Few places within the United States have the tropical or subtropical climate that can sustain fruit production of commercial-level passion fruit. Studies suggested that ethephon and ethylene could be used to accelerate ripening in passion fruit and improve postharvest quality, thus allowing the plant to be grown as an aboveground annual in the southern United States (Arjona and Matta, 1991; Dozier et al., 1991).
Alternately, species native to temperate climates such as P. incarnata could be used in breeding for fruit production and possibly contribute cold hardiness genes in combination with other Passiflora species (Winters and Knight, 1975). Passiflora incarnata has had little attention paid to it for breeding purposes (McGuire, 1999) aside from work done over many years by Knight in Florida that eventually led to ornamental cultivar releases, but none for fruit production. Most genotypes of P. incarnata used in breeding were selected from the wild without much consideration for valuable characteristics. Because of its wide adaptation to climates and soil conditions, P. incarnata certainly contains the necessary genetic variation for breeding work (Knight, 1994; McGuire, 1999). Objectives for a breeding program to introduce commercial quality fruit from P. incarnata should include rind color, pulp flavor and color, seed size, and postharvest quality just to name a few. Vine characteristics such as disease and insect resistance, moderation of rampant growth, increased flower production and timing, and shortening of flower to harvest times are also key considerations. To establish a breeding program targeted at genetic improvement, wild, native accessions should be collected and evaluated. Private breeding companies and public individuals (King, 2000) could provide potentially useful genotypes.
Interspecific hybridizations with P. incarnata have yielded interesting genotypes, but none have been commercially successful as fruit crops (Knight, 1971, 1994). Ornamental hybrids containing P. incarnata have been more notable (Knight et al., 1995). One problem with interspecific crosses between P. incarnata and P. edulis is that the F1 hybrids are often pollen sterile (Knight, 1991). Increasing ploidy level is one method to improve genetic compatibility. Colchicine has been used to increase the ploidy level of interspecific crosses involving P. incarnata to recover fertility (Bruner, 1998; Knight, 1991). The elevation of ploidy level has not always been completely successful though, as the F1 hybrids are sometimes pollen sterile and self-incompatible (Knight, 1991, 1992). Naturally occurring tetraploid plants of P. incarnata have been reported (Lloyd, 1963), but are not common. P. incarnata is also andromonoecious, which further complicates fertility issues (May and Spears, 1988; Spears and May, 1988). Hand pollination can increase fruit size and quality (Arjona et al., 1991; Knight, 1991), but the associated cost and labor involved is problematic. Natural self-compatible individuals of P. edulis exist (McGuire, 1999) and this may be the case for P. incarnata as well, although no confirmed cases have been documented (Knight, 1991). The increase to tetraploid levels may have promise for use in breeding to develop a temperate zone passion fruit (Knight, 1991), but using diverse germplasm may uncover improved fertility that could circumvent the need for chromosome doubling.
The objective of this research was to use P. incarnata genotypes from different provenances to assess viability of controlled self-pollinations and cross-pollinations within this Passiflora species.
Materials and Methods
The study was performed in 2018, 2019, and 2021 at the Mississippi State University South Mississippi Branch Experiment Station in Poplarville, MS. Passiflora propagules were collected from various locations: Florida (FL), Illinois (IL), Mississippi (MS), Missouri (MO), and Oklahoma (OK) (Table 1). P. incarnata were obtained from wild plants, either via seed or root cutting. Thus, there was expected variability within location and among plants. Some vines were also of different ages and sizes. Therefore, the floriferousness of the vines was different, leading to only a few flowers on some vines. The number of potted vines also varied between three (MS) and eight (IL).
Passiflora incarnata vine origination, and propagule type used in breeding hybridization.
Collected seeds were submerged in hot water and allowed to sit for 24 h before sowing under intermittent mist in a greenhouse. None of the seeds had any stratification or other scarification. Root cuttings were immediately put into pots with a potting mix and slow-release fertilizer under intermittent mist. On emergence, all vines remained in the greenhouse (2018) and/or fully enclosed screened high tunnel (2018, 2019, 2021), but were removed from intermittent mist once vining habit and tendril growth started. At that stage they were then re-potted into 11.4-L pots (3-gal) pots and hand watered as needed.
Controlled hand pollinations were done once a flower had opened (Fig. 1), generally between 1 pm and 3 pm. Flowers of P. incarnata are viable for pollination just a few hours on the day that it opens. Crosses started 9 May 2018 and continued through 26 June 2018, 7 July 2019 to 8 Aug. 2019, and 1 June 2021 to 28 July 2021. An anther was removed with tweezers and pollen was placed by gently rubbing the anther against the stigma. All other anthers were removed from the flower. The greenhouse and screenhouse had no other potential pollinators inside, so covering of the flower was not necessary. Evidence of positive fertilization was observed usually within 48 h, but final determination of success was not made until fruit fully developed, ≈40 to 50 d after pollinations were made. An effort was made to perform at least 20 crosses of each combination.
Examples of flower variability among Passiflora incarnata genotypes. Row 1 from left to right: Illinois, Mississippi, and Florida. Row 2 from left to right: Oklahoma, Missouri.
Citation: HortScience 57, 8; 10.21273/HORTSCI16658-22
Flower measurements were done in 2021 after flowers had opened (Fig. 2). All flower and fruit size measurements were made with a Mitutoyo Absolute Digimatic (Mitutoyo Corp., Kawasaki, Japan) and soluble solids content was measured on a digital hand-held refractometer (3810 PAL-1; Atago, Osaka, Japan). Fruit density was determined by the formula: Fd = W/(H × D), where W = fruit weight (g), H = fruit height (mm), and D = fruit width (mm). Fruit shape was calculated as H/D as described in Md Nor et al. (2019). Fruits were collected after natural abscission from the vine.
Flower structure and labeled parts of Passiflora incarnata.
Citation: HortScience 57, 8; 10.21273/HORTSCI16658-22
Data of flower and fruit measurements were analyzed by JMP (version 12; SAS Institute, Cary, NC) using a one-way analysis of variance, and means were compared with standard error of the mean or Tukey’s honestly significant difference at the 0.05 level where appropriate.
Results and Discussion
P. incarnata has previously been reported as a self-incompatible species (Krosnick et al., 2017; May and Spears, 1988). Attempts to hand self-pollinate flowers in this study validated those previous reports. Of the 122 flowers across the five P. incarnata genotypes from differing locations, none of them produced a fruit or had any indication of successful or partially successful fertilization when selfed (Table 2). If self-compatibility does exist in nature, it is likely to be rare. However, certain combinations of P. incarnata from different locations displayed combinations that produced successful fruiting, including IL × MO (52% success), FL × MO (85%), FL × OK (80%), MS × OK (40%), MO × IL (50%), MO × OK (40%), and OK × MO (80%). Interestingly, IL failed to produce a single fruit with MS and FL (Table 2). MO was moderately successful as the receiver of pollen overall, with 48% success. That is 23% greater than the overall average. This could be an indicator of more compatible pollen across a wide range of P. incarnata, especially compared with vines from IL or MS (Table 2). The best female was FL with 44% success, 19% greater than the overall average. Therefore, vine selections for pollen viability and intraspecific affinity should be a consideration for breeding of P. incarnata.
Number and percentage of successful and unsuccessful pollination events of potted Passiflora incarnata vines from five locations, crossed and self-pollinated, over 3 years (2018, 2019, and 2021).
The differences displayed across provenances show that incompatibility exists within P. incarnata but can depend on location. Hand pollinations at a single location in Florida performed by May and Spears (1988) yielded low percentages of flowers that successfully set fruit, between 13% and 20%. These percentages are similar to the overall average fruiting success for P. incarnata in this study (Table 2); however, seasonal timing did play a role in pollination success in other studies, with flowers earlier in the season tending to have higher percentages of fruiting (May and Spears, 1988; McGuire, 1998).
Tague and Foré (2005) found that most of the genetic variability of surveyed P. incarnata in South Carolina was from within a patch of vines and little variation occurred among different patches. Yet, based on the pollination results of the present study, variation among patches (provenances) must exist, as some provenances displayed a greater ability to set successful pollinations. This deserves more investigation as to why this occurs.
In total, the most fruit was collected from FL vines (45; 44%) and the least from MS vines (14; 14%) (Table 2). Despite controlled crossing under the similar conditions, IL was a poor maternal parent (Table 3). This shows that larger flower size does not equate to fecundity. Despite small flower size, FL fruit were the heaviest and had the most fruit set by far. MO had the lightest fruit and the second most fruit set. Fruit heights were significantly different, with IL and OK producing the tallest fruit. IL also produced the widest fruit; however, there were only three fruits produced, compared with all other locations that had much more fruit. FL had the fruit with the highest soluble solids level. This might indicate a breeding advantage in using FL vines as a parent to enhance fruit sweetness. OK vines were able to ripen fruit sooner than the other locations and was similar to MS. Reasons for this are not clear. MO, IL, and FL were all at or longer than 47 d (Table 3). Reducing the time from pollination to harvest would be advantageous to growers.
Fruit characteristics by female parent in intraspecific Passiflora incarnata crosses performed in 2021 among five provenances.
Fruit shape is calculated as fruit height divided by fruit width. Round fruit will have a value of 1.0, whereas more oblong fruit will be greater than 1.0, and wider, but shorter, fruit will be less than 1.0. MO fruit was the most oblong, and IL and MS the closest to a round shape. With such variability in shape, there should be possibility to breed for desirable shapes. Fruit density was calculated as fruit weight divided by fruit height times fruit width. Often, maypop (and other passion fruit species) can look to be ripening normally and of appropriate size, but on further destructive examination are completely hollow or only possess a few seeded arils. Thus, fruit density gives a measure of fullness of the fruit. The higher the value, the more packed the fruit will be with aril filled seeds. McGuire (1998) used a 1 to 5 rating system, whereas Senter et al. (1993) assessed it on a basis of seed number. A single, nondestructive measure incorporating fruit size and weight would give researchers a metric on which to compare within a species or cross. Once enough data are collected, then acceptability parameters could be implemented for use in breeding or other activity. In this limited sample, the values of fruit density ranged from 1.11 to 1.84 (Tables 3–5). The greater the value, the more densely packed the fruit will be. MS had the best filled fruit, followed by FL. Crossing of these two may be advantageous to increase fruit density. There is no established threshold for fruit density, but based on this study, 1.5 may be a cutoff where below that fruit is not densely packed enough with arils to be adequate for fruit production.
Fruit characteristics by male parent in intraspecific Passiflora incarnata crosses performed in 2021 among five provenances.
Fruit characteristics of intraspecific Passiflora incarnata crosses performed in 2021 among five provenances.
Consumers desire a consistent size and shape (Md Nor et al., 2019), which is necessary for the highest classification (Codex Alimentarius, 2014). All fruit in this study could be classified as oblong spheroid, as they were above 1.05 (Md Nor et al., 2019).
When examining the male (pollen) parent within the crosses, the number of fruits collected were different (Table 4). FL pollen produced the fewest number of fruits, with OK and MO being the most. Fruits produced from OK pollen were the heaviest, the tallest, and the widest. Although as a female parent OK fruit was not impressive, using pollen from OK anthers onto flower stigmas from other locations produced the best-looking fruit. Fruit height, °Brix, days from pollination to harvest, and fruit shape were not significantly different among the locations (Table 4).
Overall, fruit weight, fruit size, and soluble solids content measured in this study were similar to and, in some cases, greater than those previously reported (Arjona et al., 1991; McGuire, 1998; Senter et al., 1993). Hand pollinations were done by McGuire (1998), whereas Senter et al. (1993) collected insect-pollinated fruit in the wild and Arjona et al. (1991) did both. Arjona et al. (1991) reported greenhouse-grown fruit at 15.1 g and wild fruit at 38.5 g. The latter value is in line with what was observed in this study; however, the former is much lower. Maypop fruits collected by Senter et al. (1993) averaged 34.3 g and 36.7 g, both similar to what was found in the present study. Fruit weight was not reported by McGuire (1998), but fruit length (height) and fruit width were somewhat smaller, with few exceptions, than observed in Tables 3, 4, and 5. Sugar contents of fruit collected by Arjona et al. (1991) and Senter et al. (1993) were substantially less than the range in this study. However, important distinctions should be made. The cited studies were done 25 to 30 years ago under different climate conditions, in different locations, and with different maypop selections. All of these factors would likely impact the comparative results. Even so, the differences help to illustrate the diversity within P. incarnata and the still-untapped potential for breeding improvements.
All floral measurements were significantly different among locations (Table 6). IL and OK had the longest sepal length and IL had the widest sepals. OK had the longest petals and IL the widest. Both also had the longest corona filaments. These measurements indicate larger flower size. Visually, OK and IL flowers are substantially larger than flowers of FL vines. It is unknown if size equates to more insect visitation or other advantage in P. incarnata, but larger flowers often produce more, yet less concentrated, nectar (Krosnick et al., 2017).
Floral components of Passiflora incarnata from five provenances.
Further measurement of reproductive components within the flowers revealed significant differences among the locations (Table 7). MS had the shortest ovary, and IL and OK had the widest ovaries. The larger size of the ovary for IL and OK flowers is in line with the overall larger size of the other floral components, as described in Table 6. IL also had the widest stigma. FL and MS had the shortest styles. OK had the longest anthers and was similar to IL for anther width. Interestingly, despite its overall larger flower size, OK had the shortest filament, along with MS. Androgynophore length (height) was highest for IL and OK, with OK having the widest androgynophore. Androgynophore size plays a role in pollination, as taller size means that a larger insect is required to ensure adequate transfer of pollen. It may be a breeding benefit to select for shorter androgynophore length to potentially diversify the types of insects that can act as pollinators. All measurements were similar to those reported by Krosnick et al. (2017) on P. incarnata in Tennessee.
Reproductive components of Passiflora incarnata flowers from five provenances.
Self-incompatibility is separated into two divisions, heteromorphic, a physical difference that creates a barrier for pollination, and homomorphic, a condition often controlled by the multiallelic S-locus (Madureira et al., 2014). Homomorphic self-incompatibility is further divided into two groups: gametophytic and sporophytic (de Nettancourt, 1997). Previous work in Passiflora has shown self-incompatibility to be homomorphic sporophytic (Bruckner et al., 1995) and gametophytic (Suassuna et al., 2003). Although P. incarnata and P. edulis var. flavicarpa are both considered self-incompatible, there is precedent to show hybridization within a self-incompatible species has resulted in some fertile progeny (Bruckner et al., 1995; Rego et al., 1999; Suassuna et al., 2003). However, the problem is complex, and there is much more to learn about how Passiflora species, especially P. incarnata, function. In some instances, the ovary will grow to a normal size, not develop seeds, and the interior will be completely empty. This is a result of incompatible pollen (Amela Garcia and Hoc, 2011), which was not common in this study, but has been observed by the author in wild vines and in other controlled crosses.
There is still much to be learned about the potential of P. incarnata in breeding. Although cold hardiness is one of the primary reasons to use it, other important traits could come to the fore, such as disease tolerance, fruit quality characteristics (i.e., thin rind, high sugar content, lower acidity), and easier vine management compared with P. edulis because of smaller vine sizes. Growers desire a self-fertile vine to reduce the need for hand pollination and to improve overall pollination success. Presently, many passion fruit vines are grown from seed, thus not clonal, or self-propagated. Using multiple cultivars for pollination is common in other fruit crops, such as rabbiteye blueberry (Vaccinium virgatum Aiton), pecan (Carya illinoinensis Wangenh. K. Koch), and muscadine (Vitis rotundifolia Michx.), and could be done with passion fruit as long as sufficient pollinator species are present.
Much of the U.S. domestic market is not familiar with passion fruit, especially as a table fresh product. This could be a barrier to adoption, but it could also prove to be an opportunity to create a niche within the present market and expand it. Although maypop fruit quality is not equal to that of P. edulis, selecting superior wild genotypes with desirable attributes to be used in future intra- and interspecific breeding is possible based on the results of this study.
Literature Cited
Amela Garcia, M.T. & Hoc, P.S. 2011 Pollination mechanisms in Passiflora species: The common and the rare flowers – ecological aspects and implications for horticulture 23 Raskin, N.D. & Vuturro, P.T. Pollination: Mechanisms, ecology and agricultural advances. Nova Science Publishers Hauppauge, NY
Arjona, H.E. & Matta, F.B. 1991 Postharvest quality of passion fruit as influenced by harvest time and ethylene treatment HortScience 26 1297 1298 https://doi.org/10.21273/HORTSCI.26.10.1297
Arjona, H.E., Matta, F.B. & Garner, J.O. Jr 1991 Growth and composition of passion fruit (Passiflora edulis) and maypop (P. incarnata) HortScience 26 921 923 https://doi.org/10.21273/HORTSCI.26.7.921
Asch, D.L. & Hart, J.P. 2004 Crop domestication in prehistoric eastern North America 314 319 Goodman, R.M. Encyclopedia of plant and crop science. Marcel Dekker Inc. New York, NY DOI: 10.1081/E-EPCS-120017103
Bruckner, C.H., Casali, V.W.D., Moraes, C.F., Regazzi, A.J. & Silva, E.A.M. 1995 Self-incompatibility in passion fruit (Passiflora edulis Sims) Acta Hort. 370 45 52 https://doi.org/10.17660/ActaHortic.1995.370.7
Bruner, W.R. 1998 Tetraploid Passiflora incarnata Passiflora 8 1 1 4
Codex Alimentarius 2014 Standard for passion fruit codex stan 316-2014 https://www.fao.org/fao-who-codexalimentarius/sh-proxy/fr/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcodex%252FStandards%252FCXS%2B316-2014%252FCXS_316e_2014.pdf
Dai, C. & Galloway, L.F. 2011 Do dichogamy and herkogamy reduce sexual interference in a self-incompatible species? Funct. Ecol. 25 271 278 https://doi.org/10.1111/j.1365-2435.2010.01795.x
Dai, C. & Galloway, L.F. 2012 Male flowers are better fathers than hermaphroditic flowers in andromonoecious Passiflora incarnata New Phytol. 193 787 796 https://doi.org/10.1111/j.1469-8137.2011.03966.x
de Nettancourt, D. 1997 Incompatibility in angiosperms Sex. Plant Reprod. 10 185 199 https://doi.org/10.1007/s004970050087
Dhawan, K., Kumar, R., Kumar, S. & Sharma, A. 2001 Correct identification of Passiflora incarnata Linn., a promising herbal anxiolytic and sedative J. Med. Food 4 137 144 https://doi.org/10.1089/109662001753165710
Dhawan, K., Dhawan, S. & Sharma, A. 2004 Passiflora: A review update J. Ethnopharmacol. 94 1 23 https://doi.org/10.1016/j.jep.2004.02.023
Dozier, W.A. Jr, Rodriguez-Kabana, R., Caylor, A.W., Himelrick, D.G., McDaniel, N.R. & McGuire, J.A. 1991 Ethephon hastens maturity of passionfruit grown as an annual in a temperate zone HortScience 26 146 147 https://doi.org/10.21273/HORTSCI.26.2.146
Fisher, A.A., Purcell, P. & Le Couteur, D.G. 2000 Toxicity of Passiflora incarnata L Clin. Toxicol. 38 63 66 https://doi.org/10.1081/CLT-100100919
Gremillion, K.J. 1989 The development of a mutualistic relationship between humans and maypops (Passiflora incarnata L.) in the southeastern United States J. Ethnobiol. 9 135 155
Hermsen, E.J. 2021 Review of the fossil record of Passiflora, with a description of new seeds from the Pliocene gray fossil site, Tennessee, USA Int. J. Plant Sci. 182 533 550 https://doi.org/10.1086/714282
Hoch, J.H. 1934 The legend and history of Passiflora Am. J. Pharm. 5 166 170
Jorgensen, P.M., Lawesson, J.E. & Holm-Nielsen, L.B. 1984 A guide to collecting passionflowers Ann. Mo. Bot. Gard. 71 1172 1174 https://doi.org/10.2307/2399250
Killip, E.P. 1938 The American species of Passifloraceae. Field Museum of Natural History Botanical Series 19 1 613
King, L.A. 2000 The Passiflora hybrid P. ‘Excel’: P. edulis × P. caerulea Passiflora 10 2 16 18
Knight, R.J. Jr 1971 Breeding for cold hardiness in subtropical fruits HortScience 6 157 160
Knight, R.J. Jr, Payne, J.A., Schnell, R.J. & Amis, A.A. 1995 ‘Byron Beauty’, an ornamental passion vine for the temperate zone HortScience 30 1112 https://doi.org/10.21273/HORTSCI.30.5.1112
Knight, R.J. Jr 1991 Development of tetraploid hybrid passion fruit clones with potential for the north temperate zone HortScience 26 1541 1543 https://doi.org/10.21273/HORTSCI.26.12.1541
Knight, R.J. Jr 1992 Characters needed for commercially successful passion fruit Proc. Annu. Meet. Fla. State Hort. Soc. 105 280 282
Knight, R.J. Jr 1994 Problems and opportunities in passion fruit culture and development Fruit Var. J. 48 159 162
Krosnick, S.E., Perkin, J.S., Schroeder, T.S., Campbell, L.G., Jackson, E.B., Maynord, S.C., Waters, C.G. & Mitchell, J.S. 2017 New insights into floral morph variation in Passiflora incarnata L. (Passifloraceae) in Tennessee, U.S.A Flora 236-237 115 125 https://doi.org/10.1016/j.flora.2017.10.005
Lloyd, R.M. 1963 Tetraploid Passiflora incarnata in North Carolina Rhodora 65 761 79 80
Madureira, H.C., Pereira, T.N.S., Da Cunha, M., Klein, D.E., de Oliveira, M.V.V., de Mattos, L. & de Souza Filho, G.A. 2014 Self-incompatibility in passion fruit: Cellular responses in incompatible pollinations Biologia 69 574 584 https://doi.org/10.2478/s11756-014-0353-0
May, P.G. & Spears, E.E. Jr 1988 Andromonoecy and variation in phenotypic gender of Passiflora incarnata (Passifloraceae) Amer. J. Bot. 75 1830 1841 https://doi.org/10.2307/2444737
McGuire, C.M. 1998 Field performance and phenotypic variation of Passiflora incarnata L. in New York State HortScience 33 240 241 https://doi.org/10.21273/HORTSCI.33.2.240
McGuire, C.M. 1999 Passiflora incarnata (Passifloraceae): A new fruit crop Econ. Bot. 53 161 176 https://doi.org/10.1007/BF02866495
Md Nor, S., Ding, P. & Sea, C. 2019 Role of assisted pollination in fruit shape of purple passion fruit (Passiflora edulis Sims.) Intl. J. Agr. For. Plantat. 8:128–133. https://ijafp.com/wp-content/uploads/2019/09/KLIAFP8_071.pdf
Rego, M.M., Rego, E.R., Bruckner, C.H., Silva, E.A.M., Finger, F.L. & Pereira, K.J.C. 1999 Self-incompatibility in passionfruit: Evidence of two locus genetic control Theor. Appl. Genet. 98 564 568 https://doi.org/10.1007/s001220051105
Senter, S.D., Payne, J.A., Knight, R.J. & Amis, A.A. 1993 Yield and quality of juice from passion fruit (Passiflora edulis), maypops (P incarnata) and tetraploid passion fruit hybrids (P edulis × P incarnata) J. Sci. Food Agr. 62 67 70 https://doi.org/10.1002/jsfa.2740620109
Spears, E.E. Jr & May, P.G. 1988 Effect of defoliation on gender expression and fruit set in Passiflora incarnata Amer. J. Bot. 75 1842 1847 https://doi.org/10.2307/2444738
Suassuna, T. de M., Bruckner, C.H., de Carvalho, C.R. & Borem, A. 2003 Self-incompatibility in passionfruit: Evidence of gametophytic-sporophytic control Theor. Appl. Genet. 106 298 302 https://doi.org/10.1007/s00122-002-1103-1
Tague, R.T. & Foré, S.A. 2005 Analysis of the spatial genetic structure of Passiflora incarnata in recently disturbed sites Can. J. Bot. 83 420 426 https://doi.org/10.1139/b05-014
Wehtje, G., Reed, R.B. & Dute, R.R. 1985 Reproductive biology and herbicidal sensitivity of maypop passionflower (Passiflora incarnata) Weed Sci. 33 484 490 https://doi.org/10.1017/S0043174500082709
Winters, H.F. & Knight, R.J. Jr 1975 Selecting and breeding hardy passionflowers Amer. Hort. 54 22 27
