Despite its crucial role in fertility, relatively little has been published on the human endometrial stroma. The first experiment reported in this thesis was designed to provide quantitative baseline data on the human stroma during the mid to late luteal phase, when it plays a major part in normal pregnancy. Subsequent chapters investigate the effects of an antioestrogen (clomiphene citrate-CC); growth hormone (GH) supplementation of patients on hormone replacement therapy (HRT) due to lack of endogenous ovarian steroids; and patients with unexplained recurrent miscarriage. Finally a novel 3-dimensional in vitro model of human stroma is described along with the effects of steroid supplementation on stromal cells grown in the model. In all cases of in vivo studies, conventional stereological methods were used to obtain quantitative morphological data from at least 6 subjects per group, from both control (fertile) and experimental (infertile) subjects using light and electron microscopy. In addition, several staining techniques were also used to demonstrate qualitative changes that occur in human endometrial stroma. During the mid to late luteal phase, endometrial stroma and blood vessels underwent substantial changes, and thus quantitative and qualitative baseline data have been established to fill a gap in this important area of reproductive biology. CC caused no substantial changes in stromal structure in women of proven fertility and therefore its advantageous effects on ovulation are not negated at the level of the endometrium. GH supplementation had no effects on infertile human endometrial stroma from subjects on HRT, (in either premature ovarian failure or Turner's syndrome groups). However, both infertile groups had endometrial stroma which significantly differed from matched fertile endometrium, suggesting an impaired endometrial development in infertile subjects which was not reversed by HRT, either with or without Gil treatment. Stromal morphology did not differ between the recurrent miscarriage groups, however data suggested the existence of several subgroups which made firm conclusions difficult. The tissue culture model provided preliminary data suggesting it to be a potentially very useful technique for the study of stromal cell biology.