The main production process of species occurring in the coma of comets is t
he photodestruction of molecules initially present in the nucleus ices and
non-refractory grains or trapped inside the nucleus "material". Grains can
also be a source of molecules in the coma. Chemical reactions may occur bet
ween coma species. Consequently, although chances that an abundant coma spe
cies has not been detected are now small, the coma composition is certainly
quite different from that of the nucleus. Except For the molecules release
d directly at the nucleus surface, all coma species are produced in an "ext
ended region" or come from ia distributed source" Since the recent literatu
re is rich in reports on observations of molecules and species possibly not
initially present in the comet ices or not released at the nucleus, a gene
ral discussion of how coma species are stored, processed or produced is pre
sented, based mostly on observational results. What is at stake is the prop
er modeling of the coma structure, hence an accurate derivation of the nucl
eus composition from coma observations.