LAUSR

ABSTRACT Araucaria angustifolia is the only native conifer of economic importance in Brazil for wood and nut production. There are several constraints to the genetic improvement of the species. Vegetative ramets display strong and lasting plagiotropism, which is difficult to pre-select for, and it is a dioecious species. Trees reach reproductive maturity at 12–15 years old and there is currently no method to distinguish the sex of the plant before this time. Early detection of plagiotropism or early identification of plant sex would improve efficiencies in breeding programs and seed orchard establishment. Therefore, we investigate the potential of near infrared (NIR) spectroscopy analysis as a tool for predicting growth habit and gender of A. angustifolia. Spectra from needles at the stem apex were taken from 12-month-old plants for growth habit calibration using the presence or absence of a lateral shoot to distinguish between future growth habits. Model validation was done on needles from the apical shoots of orthotropic trunks and from clones displaying plagiotropic growing habit. For plant sex, spectra were collected on wood from bark windows and from multiple needles located midway up the main stem. Needle spectra from individual plants were averaged to produce a single spectrum per plant. Multivariate analysis was performed on spectra and associated reference data. Results indicate that NIR has the potential to predict sex and growth habit and may be a useful screening tool in improvement of Araucaria.