Fluctuating temperatures are characteristic of the en-1;ronment of most organisms, yet constant temperatures .~e most ~requently .used for laboratory st_udies on th~ir ''ie histones. The hterature has been revtewed to pomt :;ut some of the ways in which constant-temperature data ;, 1 r rates of development, developmental thresholds, fer-;ility and other life-history phenomena could be mis-:,adi~g when an ecologist attempts to model a natural ;,stem or to extrapolate to field conditions. To imple-:;1ent the replacement of constant-temperature studies .;ith fluctuating temperature studies, 2 previously re-?orted timer-contrail~ bimetallic or m~rcury thermostat Jevices and a new timer-controlled sohd-state controller