The aim of this study was to determine the composition and microbiological quality of honeybeecollected pollen from various botanical and geographical origins of Spain and to define the following physicochemical and microbiological characteristics: water content, water activity (A w ), mean pollen pellet size, protein content, fat content, fatty acid composition, free amino acid distribution, sugar spectrum, mineral elements, dietary fiber, starch content, aflatoxins, and microbiological parameters. The free amino acid spectrum showed a high level of proline (63.1%) of the total free amino acid content (x ) 31.6 ( 4 mg/g). The main sugars identified by gas chromatography were fructose, glucose, and sucrose, with small amounts of di-and trisaccharides. A high level of dietary fiber was detected (x ) 13.7 ( 1.3 g/100 g). The main fatty acids detected corresponded to C 18:2 , C 18:3 , and C 18:1 acids. The mineral elements showed a predominance of potassium, phosphorus, calcium, and magnesium. The pollen of Cistus ladaniferus was detected as the dominant pollen. Among the microbiological parameters, a high number of molds, total aerobic counts, and the presence of coliforms and Streptococci "D" of Lancefield were found. Aflatoxins were not detected.

The composition and bacteriostatic activities of fifteen propolis samples from various botanic and geographic origins were determined. Twenty‐six phenolic components were identified by high‐performance liquid chromatography with array photodiode detector. Acacetin and apigenin were most abundant. Pinocembrin, quercetin, rutin and vanillin were present in lesser quantities. Variance analysis shows significant differences (P≤0.05) in the contents of phenols, flavonoids and active components. The minimum inhibitory concentration of propolis is about 53 times higher than that reported for tetracycline againstBacillus subtilis andStaphylococcus aureus, and about 400 times higher againstEscherichia coli.

The ripening process of kiwifnits (Actinidia deliciosa), was studied to determine the best harvesting time and storage potential for commercial shipping. The samples were taken from two representative interior and coastal areas of "Maresrne " (Catalonia) with known and controlled agroclimatic features. Relative humidity and temperature conditions were controlled for cold storage. The samples for analyses were taken during the maturation phase of the plant (52 days) as well as in cold storage (182 days). Soluble solids (SS)concentration, flesh 3rmness. sugar composition, total acidity and vitamin C content were measured, the first three parameters were indicative of the degree of ripeness. At the moment of reaching the peak concentration of sugar and SS (optimal eating time) the protein, lipid, content of mineral elements (K, Fe, Mg and Ca) and the parameters specijied above were determined.

Mineral composition, vitamins (group B, C and D), polyphenols, purine alkaloids, sugars and phytic acid were determined in twelve samples of cocoa husk of the main varieties of cocoa beans ( Theobroma cacao L.) cultivated in Africa and South America. Macro- (K, Mg, Ca, Na, P, and Na) and micro elements (Cu, Zn, Fe, and Cr) were quantified using an inductively coupled argon plasma-optical emission spectrometer (ICP-OES), on dry-ashed samples. Vitamins, phenolic compounds, alkaloids, sugars, and phytic acid were determined using HPLC. The mineral and vitamins intake that cocoa husk consumption could represent was evaluated with regard to the Recommended Dietary Allowances (RDA). The percentage of phytic acid is medium (0.98±0.13 g/100 g). The most abundant phenolic compounds found were caffeic acid and gentisic acid (2,5-dihydroxybenzoic acid).

The number of mesophilic aerobic colonies was determined in 72 samples of mono- and multifloral honey from various sources by the plate count and the membrane filter methods. The presence of motile colonies made the plate counts unreliable. The microorganism producing these colonies was identified as Bacillus alvei. Colony counts could only be carried out in 27 of the samples when using the plate count method, while with the membrane filter method the number of colonies was counted in all the samples.