This paper uses three methods to estimate quality option values for CBOT Treasury bond futures contracts. It presents evidence regarding: (1) payoffs from exercising this option at delivery, (2) estimates from a T-bond futures pricing model that incorporates this option, and (3) estimates obtained from an exchange option pricing formula. The results indicate that this option is worth considerably less than reported by Kane and Marcus (1986a). For example, payoffs obtained by switching from the bond cheapest to deliver three months prior to delivery to the one cheapest at time of delivery average less than 0.30 percentage points of par.