A passion for passion flowers

Gerhard Prenner, researcher in plant morphology and anatomy, presents his recent ontogenetic study on one of the passion flowers, Passiflora lobata. He reveals insights about its peculiar flowers and highlights the Easter connection of the genus Passiflora.

Gerhard Prenner

By Dr Gerhard Prenner

Passiflora or passion flower is a genus of more than 550 species which are members of the Passifloraceae family. The genus, which contains vines, lianas, shrubs and trees, is predominantly distributed throughout Mexico and Central and South America, with a few species endemic to Southeast Asia and the Pacific. Because of their decorative flowers some species are frequently cultivated as ornamentals and, even better, the fruits of some passifloras are delicious and are traded as passionfruit or maracujá (P.edulis), sweet grenadilla (P. ligularis), and other more locally grown species. 

The name 'passion flower' refers to the passion of Jesus and the genus therefore has a particular relevance at Easter. Spanish Christian missionaries adopted the unique structure of the plant as symbols of the last days of Jesus and especially his crucifixion. In explaining these symbols I can highlight some of the key morphological features of flowers of the genus Passiflora and describe why the genus is a fascinating subject for my research on floral development. The symbolism of the different floral parts is as follows:

  • the five petals and five sepals taken together represent the ten faithful apostles (excluding St Peter the denier and Judas the betrayer) 
  • the flower's radial filaments or corona represent the crown of thorns 
  • the three stigmas represent the three nails and the five anthers the five wounds (four by nails and one by the lance) 
  • the tendrils represent the whips used in the flagellation of Christ
Labelled diagram of Passiflora phoenicea floral structures
Standard floral structure of a passionflower © Gerhard Prenner

Passiflora phoenicea is a radially symmetric flower with five sepals (s) and five petals (p), both coloured red, a ring of radial filaments (the corona), five stamens (yellow; one highlighted) and an ovary composed of three carpels (one style highlighted).  

This floral structure, or groundplan, of Passiflora occurs in most species in the genus but there are always exceptions to the rule and it is often the exceptions that provide the greatest insights. Passiflora lobata, the subject of my recent research on floral development, deviates from the standard floral groundplan of passion flowers in that it has four styles (its pistil is formed of four carpels) instead of three. Furthermore, its flowers have a single plane of symmetry (they are zygomorphic) instead of the many planes of symmetry (radial symmetry or actinomorphy) commonly found in the flowers of most other passifloras (e.g. MacDougal 1994; Bernhard 1999; Krosnick et al. 2006, 2013). 

Passiflora lobata is a vine up to 10 m long, with three-lobed leaves and distinct extrafloral nectaries at the lower half of their flower stalks (pedicels). It occurs in moist to wet tropical to lower montane forests from south-eastern Nicaragua, Costa Rica and Panama to northern Colombia. 

Because of its peculiar flowers Passiflora lobata was not originally classified as a passion flower and was first described as Tetrastylis lobata. However, molecular phylogenies (evolutionary trees based on DNA data) show that the genus Tetrastylis is in fact nested within other passion flowers and the species has therefore been renamed Passiflora lobata (MacDougal, 1986; Muschner et al., 2003). The uncommon arrangement of the floral parts (the groundplan) prompted me to study the development (ontogeny) of the flowers and the inflorescences in detail, using Kew's scanning electron microscope (SEM) (Prenner, in press).

My study was driven by the following key questions: 

  • What is the sequence of organ initiation? 
  • Where and how is the additional carpel formed and when and how is floral zygomorphy manifested? 
  • Can ontogeny help to clarify the nature of the tendril and corona in Passifloraceae?

Answering these questions should shed additional light on the floral evolution in Passifloraceae and increase our understanding of Passiflora lobata and its evolutionary relationships. 

My work revealed some interesting and hitherto not recognised characters (Prenner, in press). Previously, the tendril was interpreted as the modified pedicel of a central flower which got lost in the course of evolution. The present study showed that tendril formation only starts after the two flanking flowers are initiated. This argues against the pedicel hypothesis where one would expect an earlier (before the two flanking flowers) formation of this structure. Late formation favours an alternative interpretation, namely that the tendril is homologous to the terminal part of the inflorescence axis. The fact that in some species flowers were found on the tendril also favours this hypothesis. The flowers are formed in pairs and show mirror symmetry, which is manifested very early in ontogeny.

The five sepals are formed in a spiral and followed by five petals, which are formed successively adjacent to each other. This is a rare pattern of organ formation and the first report in Passifloraceae. Frequently a sixth petal is initiated which never fully develops and which could be interpreted as the first outgrowth or frill of the corona (see top image). This therefore might be interpreted as derived from the perianth. 

In the above images, a sixth petal primordium (P6) is found in both flowers. Five sepals (S1–S5) are formed in a spiral and five stamens (A) arise simultaneously (sepals removed).  

The simultaneous stamen formation is likely to be a constant character in the genus. Within the pistil, which is formed of four fused carpels, the abaxial carpel (the one facing away from the inflorescence axis) is always formed in front of the first-formed sepal. The remaining three carpels alternate with the stamens. This means that a positional change took place from the typical trimerous (containing three parts) ovary with two carpels in front of stamens to only one antestaminal carpel in P. lobata. This shift might have opened up space for the formation of a fourth carpel. 

Download a high resolution file of the SEM images

The present study leads on to important questions concerning the circumstances in related taxa: 

  • How similar are the other species which were formerly described in Tetrastylis and which have a similar groundplan as P. lobata
  • How do the discovered characters relate to other tetramerous Passifloraceae, such as AncistrothyrsusMitostemma and Dilkea?

Answering these and related questions will further help to enhance our knowledge of this remarkable genus.


  • Bernhard, A. (1999). Flower structure, development, and systemtics in Passifloraceae and in Abatia (Flacourtiaceae). International Journal of Plant Sciences 160: 135–150. Available online
  • Krosnick, S. E., Harris, E. M., Freudenstein, J. V. (2006). Patterns of anomalous floral development in the Asian Passiflora (subgenus Decaloba: supersection Disemma). American Journal of Botany 93: 620–636.Available online
  • Krosnick, S. E., Porter-Utley, K. E., MacDougal, J. M., Jorgensen, P. M., McDade, L. A. (2013). New insights into the evolution of Passiflora subgensu Decaloba (Passifloraceae): phylogeentic relationships and morphological synapomorphies. Systematic Botany 38: 692–713. Available online
  • MacDougal, J. M. (1986). A new combination in Passifloraceae. Phytologia 60: 446. 
  • MacDougal, J. M. (1994). Revision of Passiflora subgenus Decaloba section Pseudodysosmia (Passifloraceae).     Systematic Botany Monographs 41: 1–446. 
  • Muschner, V. C., Lorenz, A. P., Cervi, A.C., Bonatto, S. L., Louza-Chies, T. T., Salzano, F. M., Freitas, L. B. (2003). A first molecular phylogenetic analysis of Passiflora (Passifloraceae). American Journal of Botany 90: 1229–1238. Available online
  • Prenner G. in press. Floral ontogeny in Passiflora lobata (Malpighiales, Passifloraceae) reveals a rare pattern in petal formation and provides new evidence for interpretation of the tendril and corona. Plant Systematics and Evolution. DOI 10.1007/s00606-013-0961-0 Available online

Read & watch