Thesis (PhD(Agric) (Animal Sciences))--Univsersity of Stellenbosch, 1992. / The anatomy of the muscles of the hindquarters of the ostrich is described and presented graphically for the purpose of the present and future studies in meat science involving the ostrich. In the past, the lack of an accurate anatomical description of this species has led to the incorrect identification of certain muscles by researchers. The process of growth is described for 51 components of the ostrich body, which includes 21 muscles and four organs. Gompertz functions were calculated to formulate growth models for these components. Acceptable goodness-of-fit was obtained for the formulated models regarding the raw skin area, mass of the head, heart, filet and other muscles as well as certain bones. No information on similar models could he found for the description of growth of individual muscles or body parts of other meat animals. However, similar growth models are available for the description of live mass changes over time. Since the general growth curve is sigmoidal it is difficult to fit accurate curves by arithmetical methods. This has led to the development of simpler linear models, which express the growth of a body part relative to the growth of the same entire body. Time based predictions of yield are thus impossible. In fact, many of these allometric equations are simply in a different mathematical form, rather than being totally different equations. Furthermore, the latter equations cannot be used to predict accurate values at the positive and negative extremes since unrealistic values are obtained. Computer programs are presented in the dissertation which can be used without modification on various species, such as fish, poultry, plants and forests, as well as micro organisms in plant pathology, food science and sewage. With the development in computer technology and by means of the suggested programs, it is possible to describe the growth process on a time scale with ease. It is also possible to estimate the mass of any body part at birth (and possibly at conception) as well as at maturity. Evidence is provided that unrealistic asymptotic values are obtained with time based linear and polynomial functions as well as with published non linear allometric models for the description of growth in the ostrich. A goodness-of-fit test for the Gompertz function is presented. It is advisable to carry out this test before fitting alternative models, since the relationship between two variables is only linear on a double log scale when both variables can be described by similar functions on a time scale. A method for the calculation of the standard error at the point where maximum growth occurs is also presented. These methods can successfully be used for any of the above mentioned species. The point at which maximum growth occurred was used to compare the muscles of the hindquarters of the ostrich with that of Merino rams. This showed that these two species differ markedly in this regard. Furthermore, it was proved that certain economically important muscles still gained weight after the optimal skin size (120 dm2) was reached at the age of ten months. It is also shown that the fat content of ostrich meat (filet) varied independently of the abdominal fat content, and that it reached a minimum value of 0.3% at the popular slaughter age of 14 months. Reliable methods for age estimation of a slaughtered bird is presented, based on the growth models. This information can be used to develop a classification system for ostrich meat and other meat animals.