BIOPHYSICS. For the article ''A molecular mechanism for osmolyte-induced protein stability,'' by Timothy O. Street, D. Wayne Bolen, and George D. Rose, which appeared in issue 38, September 19, 2006, of Proc Natl Acad Sci USA (103:13997-14002; first published September 12, 2006; 10.1073͞pnas. 0606236103), the authors note the following: ''For Fig. 2 of our article, we inadvertently published a plot of the contact surface area rather than the accessible surface area as intended. Also, the correlation coefficient given should be 0.81, not 0.88 as in the original figure caption. All other aspects of the article remain unaffected by this correction. We regret the errors.'' The corrected figure and legend appear below. To investigate the functions of circadian neurons, we added two strategies to the standard Drosophila behavioral genetics repertoire. The first was to express a polyglutamine-expanded neurotoxic protein (MJDtr78Q; MJD, Machado-Joseph disease) in the major timeless (tim)-expressing cells of the adult brain. These Tim-MJD flies were viable, in contrast to the use of cell-death gene expression for tim neuron inactivation. Moreover, they were more arrhythmic than flies expressing other neurotoxins and had low but detectable tim mRNA levels. The second extended standard microarray technology from fly heads to dissected fly brains. By combining the two approaches, we identified a population of Tim-MJD-affected mRNAs. Some had been previously identified as sex-specific and relevant to courtship, including mRNAs localized to brain-proximal fat-body tissue and brain courtship centers. Finally, we found a decrease in the number of neurons that expressed male-specific forms of the fruitless protein in the laterodorsal region of the brain. The decrease was not a consequence of toxic protein expression within these specialized cells but a likely effect of communication with neighboring TIM-expressing neurons. The data suggest a functional interaction between adjacent circadian and mating circuits within the fly brain, as well as an interaction between circadian circuits and brain-proximal fat body. C ircadian rhythms are widespread in nature. As a consequence, many aspects of behavior, metabolism, and physiology show cyclical changes during each 24-h period. These include locomotor activity rhythms, which are intimately connected to the Ϸ75 pairs of known clock neurons within the adult Drosophila brain (e.g., ref. 1). These cells express cycling levels of clock mRNA and protein, including products of the clock genes period and timeless (per and tim; ref. 2). The circadian molecular program also runs within several peripheral (noncentral brain) tissues, including eyes, Malpighian tubules, leg sensilla, and antennae (3). Peripheral tissues contribute to other circadian outputs such as olfaction (4) and reproductive behavior (5-7).Several tools exist to eliminate or inactivate Drosophila pacemaker neurons, including agents that cause cell death, e.g., the proapoptotic genes hid and reaper (8), or those that inacti...