This paper investigates the optimal design of weather bonds for reinsurance purposes. The motivation for this task comes from an empirical study showing that German farmers are not willing to pay the premiums for weather insurance that insurers ask for. Since reinsurance costs constitute a major cost component of insurance premiums, minimizing these costs could decrease the observed gap between the willingness to pay and the willingness to accept the cost of the insurance. Against this background, we put forth the proposal to transfer weather risk directly to the capital market by issuing weather bonds. The structure of the weather bond is optimally designed in a utility maximizing framework that involves farmers, insurers, and capital market investors. The approach is illustrated by an example of securitizing draught risk in crop production in Germany.