Abstract
Floral syndromes can be a useful tool for predicting the identity of pollinators from floral morphology. However, the reliability of floral syndromes are still debated in the literature, and can often result in effective pollinators being overlooked. Pollination by moths has historically been associated with a floral syndrome comprising white tubular flowers that are heavily scented at night, but there have been few experimental tests of this association. To test whether moths preferentially land on flowers that are white and tubular, we allowed two moth species ( Agrotis ipsilon (Hufnaggel, 1766) and Ichneutica mutans (Walker, 1857)) to choose freely between three plant species ( Vaccinium corymbosum , Ericaceae, Pittosporum tenuifolium , Pittosporaceae, and two forms of Leptospermum scoparium , Myrtaceae) which represent four different floral morphologies. Anthers on the flowers were marked with fluorescent powder dye (pollen‐tracker), with the presence of pollen‐tracker on moth bodies after the experiment indicating floral visitation. We detected no floral preferences for any of the plant species by either moth species, suggesting moths will visit flowers that lack features associated with the syndrome even when in the presence of white, tubular flowers which might be predicted to be more attractive to moths. We also examined moth bodies to determine where pollen‐tracker was carried most often and if this varied among the floral types. Our results indicate that some moth body parts may be more important than others for pollen transfer, depending on the morphology of the flowers visited. After visiting tubular flowers ( Vaccinium and Pittosporum ), pollen‐tracker was most often located on the proboscis, but after visiting open‐access flowers ( Leptospermum ) pollen‐tracker was most often located on the legs. Future studies looking at the presence of pollen on moth bodies should consider this and ensure the entire moth body is examined for the presence of pollen.