For 30 years there has been experimental work aimed at transplanting islets for the treatment of diabetes with a view to curing the disease and preventing the secondary complications. Many technical difficulties were experienced, first in isolating the islets without damaging them, and second in finding a suitable place to inject them, but until recently the results of a vascularized pancreas transplant have been superior to islet transplantation. In 2000, the group in Edmonton, headed by Shapiro, published encouraging results using a different immunosuppression in transplanting patients earlier in the course of their disease than had been attempted previously. The results were excellent at a year and good at 2 years in patients with Type I diabetes, however there was the rather worrying attrition at five years. Nevertheless, the Edmonton observations were proof of concept and have intensified interest in treating diabetes and other diseases where a specific protein synthesis was required by cell transplantation and/or genetic engineering. The recent interest in embryonic stem cells extenuated these efforts and progress is being made in defining the difficulties, which are greater than most workers would have predicted. In this review, the subject is discussed explaining where progress needs to be made in order to provide treatment that would be of value to patients.