This study responds to the need to further the understanding of persistence of females in science. By studying college seniors it provides data on a population that previously has not been studied well. By studying successful graduates (3.00-4.00 cumulative GPA) it represents a group of females most likely to persist in science beyond the bachelor's degree. Other criteria for entry into the research population required females to be: American educated; graduates of Iowa State University from 1986-1994; and majors of a laboratory-based science;Surveys were sent to 221 females initially meeting these criteria. The 165 returned surveys that were deemed usable constituted a return of 75 percent;Sixty-eight percent (n = 112) of respondents persisted in science beyond the bachelor's degree to pursue either advanced graduate studies or professional programs of study. The most important factor related to persistence was receiving encouragement from others. A "chilly classroom climate" (nonstimulating and/or sexist environment) was considered to be the most important factor related to experiencing difficulties during this time. T-tests were used to determine if the two groups of persisters differed on six factors related to undergraduate experiences. Persisters differed on only one factor. Females pursuing advanced graduate studies had significantly more positive attitudes toward laboratory work than females pursuing professional programs of study;Nonpersisters considered financial problems, obtaining a job, role conflicts, and lack of information regarding what to study to be the most important factors related to their nonpersistence. If access was improved (through part-time study and video technology) they believed participation in science could increase;Six factors were used to study undergraduate science experiences. Two-way ANOVA testing resulted in significant differences being found between females who graduated "with distinction" (3.50-4.00 GPA) and those who did not (3.00-3.49 GPA) on their attitudes toward relationships with science professors, self-confidence for science studies, and academic advising experiences. When the six factors were used in multiple regression analysis to predict persistence in science after graduation, two factors emerged as predictors--self-confidence for science studies and the masculine nature of the science classroom. These factors, however, explained only a small amount of the variance.