Mathematical models of cotton fiber length distribution under various breakage models are discussed. Theoretical analyses are given, including the demonstration of certain factors, capable of measurement. which afford the possibility for numerical con firmation of the respective models. For example, the model for random fiber breakage independent of length shows that where μ F and μ G are the means and γ F and γ G are the absolute second moments about the origin of the parent and daughter populations, respectively. In addition, mathematical models, somewhat more complicated, are given when breakage is proportional to length and for the case where the fibers are held and then subjected to breakage.

Current methods of evaluating the amount and particle size of trash in ginned cotton are reviewed and the need for simplicity and speed in obtaining a measurement of trash content is evidenced. The authors discuss a new optical-electronic method of scanning a classer's sample to record both particle size and number of particles. The sum of the areas in the sample which are darker than the background is taken as a percent of the total area scanned. The method utilizes a closed circuit television camera, which in a fraction of a second traverses the face of a sample and reads out the desired information directly. Results of tests designed to prove the reliability and validity of the method are shown and a comparison to current methods of evaluation is given. Diagrams of the optical and computing systems used in the method are explained in detail.

An experiment was designed as an instructional method, elective class participation (ECP), in which class attendance was not required and the instructor tended to be a guide rather than lecturer. Ss were adult students enrolled in one of two semesters of introductory psychology. For two years data were collected on all Ss. Half the sessions were taught the conventional way; the other half used the ECP approach. Hypotheses tested were: Ss in the ECP groups would show more posttest gain over the pretest scores than those in the control groups; the ECP method would elicit favorable reactions. The criterion measure was posttest gain over pretest score. Biserial part correlations were computed between ECP and control groups and posttest score residualized with respect to six control variables: sex, hours attempted, semester grade point average (GPA), cumulative hours completed, cumulative GPA, and pretest score in course subject matter. Four correlations were positive, indicating that enrollment in the ECP group was related to a higher posttest score. In one group the relationship was statistically significant. The data were further analyzed into the following groups: ratings of classes which consisted primarily of lecture, demonstrations, films, and feedback quizzes. No significant differences were found among ratings of the last three, but when each of the last three was compared with the lecture classification, it was found that Ss rated ECP classes higher than the locture classes. It is apparent that Ss preferred to attend (or at least rated the quality higher) classes in which novel techniques were used. Ss' ratings of the ECP course indicated that 59% rated it average or better and a satisfying learning experience while 41% rated it fair or poor. KINGS1.EY M. W'tt:n~rc:r.