The absence of CO from the dust peak around ε Eri

We have searched for the J = 2-1 emission of carbon monoxide (CO) from the brightest peak in the epsilon Eridani (epsilon Eri) dust disc - without success. The non-detection sets 3sigma upper limits to the mass of CO gas in the dust peak of 3.3 x 10(16) kg and to the total amount of (H-2) gas in the disc of 8 x 10(19) kg. The paucity of gas in the disc (there is similar to1000x more dust) constrains the composition of orbiting comets or icy planetesimals. If these comprise the pre-eminent reservoir of dust and gas in both the Edgeworth-Kuiper Belt (EKB) and the epsilon Eri disc then it seems unlikely that the epsilon Eri planetesimals can be as rich in volatiles as their EKB counterparts.