degradation. As stated previously, color of the host plant plays a maj or role in whether or not a female butterfly is successful in finding an acceptable host (Prokopy & Owens 1983). As stated by Myers (1985) and Wolfson (1980), color may depend on plant chemistry, which is directly related to the physiological condition of the plant. It is possible that as the experiment progressed, the physiological condition of the plants gradually rendered them less and less visible, attractive, and suitable to gravid female butterflies. Conspecific egg loads have been found to be oviposition deterrents for many species of butterflies, including those in the genus Danaus (Rausher 1979b, Rothschild & Schoonhoven 1977). For this very reason, eggs were removed daily in an effort to minimize oviposition deterrence. However, some species of Lepidoptera have been known to chemically detect the presence of conspecifics (Renwick & Radke 1980). It is possible that residual chemicals from the egg remained on the leaf surface, which may have acted as an oviposition deterrent. Despite these possible explanations for a bell-shaped curve in oviposition for day and the very significant p-value, we must not overlook the fact that this bell-shaped curve was observed solely at Greathouse. Because Greathouse yielded significantly more data than Shady Oaks, the analysis of the combined data overemphasized the results from Greathouse. However, there was clearly a pronounced day effect at Shady Oaks as well, with fluctuations up and down throughout the 10-day trial. Analysis of the day main effect suggests that more trials need to be conducted if we are to confidently say that a bell-shaped pattern is indeed occurring. Main Effect 2: Geographic Location The effect that geographic location had on egg number was also very significant (p-value < 0.0001). When the data are analyzed by location, we see that the Greathouse trial yielded approximately three times more eggs overall than the Shady Oaks trial. Similarly, almost three