We measured responses to a large panel of odorants from a diverse family of chemical substances, including odors with a pheromonal value for bees. We found that odor-responses in mAPT glomeruli did not differ from odor-responses in lAPT neurons in terms of response probability and odor-response onset time. However, mAPT glomeruli had larger odor responses, and a slightly delayed late odor-response onset. The results are discussed with respect to other possible functions of parallel processing in find more the two olfactory subsystems. Our novel technique should allow accessing concealed and/or hidden

brain surfaces without tissue damage in other brain preparations. Standard glass coverslips (20 × 40 mm, 170 μm thick)

were gold-sputtered on one side using a standard gold-sputter for raster electron microscopy. Coverslips have an optically perfect surface, and are therefore well suited as mirror substrates. Gold sputtering is widely available and affordable, making this a good low-budget technique. The coverslips were then broken by gentle pressure with forceps, and from the fragments, pieces with appropriate size and shape were selected for the preparation. Forager honeybees were collected AZD6244 from indoor hives kept at 12:12 L:D regime, chilled until motionless, and mounted in custom made Perspex chambers (Fig. 1B). A window was cut into the head cuticle, surface trachea were removed, and the brain was bathed in a calcium dye solution (Calcium-Green 2-AM, first dissolved in Pluronic+DMSO, then in saline solution. Saline, in mM: 130 NaCl, 6 KCl, 4 MgCl2, 5 CaCl2, 160 sucrose, 25 glucose, 10 HEPES, pH = 6.7, 500 mOsm; dye, Pluronic and Ribose-5-phosphate isomerase DMSO from Molecular Probes, NL; all other chemicals from Sigma, Germany).

Incubation with the calcium dye took place at approx. 14 °C for 1 h, then the animals were placed at room temperature. For more details, see (Galizia et al., 1997 and Galizia and Vetter, 2004). The head capsule was repeatedly rinsed in fresh saline. Prior to imaging, a mirror was placed either lateral or medial to one of the bee’s antennal lobes, at an angle of approx. 45° (Fig. 1B), and fixed with wax to the imaging chamber. Coverslips were inserted with the glass side facing up, because this orientation gave better images. The animal was then placed into the measurement setup, and calcium measurements were started. Recordings were done using a CCD-camera based imaging system (640 × 480 pixels, TILL Photonics, Germany), with 12 bit dynamic range, through a 20× lens, NA = 0.5, with 3.3 mm working distance (Olympus, Japan). The focal plane was chosen as to either obtain a direct view of the frontal surface of the antennal lobe (Fig. 1C), or place the mirror image of the antennal lobe’s medial or lateral side into focus (Fig. 1D).