The purpose of this study was to assess a new criterion for binocular comfort analogous to the classical Sheard's criterion. Instead of equating the fusional demand with the monocular phoria as is done when Sheard's criterion is applied, the new criterion uses a calculated binocular fusional demand. The binocular demand was derived using a clinical measurement of the convergence accommodation per convergence (CA/C) ratio. Sheard's criterion was also evaluated. Other commonly used indicators of binocularity (heterophoria, vergences, accommodative amplitude, facility and response, fixation disparity, and the associated phoria) were measured. One hundred subjects (52 males, 48 females; mean age 26 years) were classified as either symptomatic or asymptomatic by an interviewing clinician. The examining clinician was intentionally masked as to the classification of the subjects. We hypothesized that the new criterion would best discriminate between the two groups of patients inasmuch as it is based on currently accepted dual-interaction models of accommodation and vergence. Our analysis confirmed that the CA/C ratio corresponded closely to those published previously (mean = 0.06 D/delta). Significant differences (p less than 0.05) were determined between the symptomatic and asymptomatic groups for gender, near phoria through a +2.00 D add, accommodative amplitude, positive vergences at near, and both the classical Sheard's and the new criterion. The new criterion was the best discriminator between the groups, identifying 72% correctly, an improvement of 6% over the classical Sheard's. However, various stepwise discriminant analysis procedures consistently failed to demonstrate that the calculated binocular fusional demand or the new criterion was superior to the near phoria or the classical Sheard's value. These results suggest potential clinical utility for new procedures based on recently described models of accommodation and vergence, but further development appears necessary.