Studies were performed to characterize the morphology and vascular reactivity of the allografted cerebral microcirculation. Cerebral cortical tissue was allografted into the cheek pouch of the hamster so that cerebral parenchymal vessels could be studied. The vascular morphology was characterized by a large number of looping vessels. The ultrastructural examination indicated viable cerebral tissue containing typical vessels, that is, "tight" junctions, not like those of the cheek pouch. Also, the microvasculature was impermeable to 150, 70, and 20 kDa fluorescein isothiocyanate dextrans. Angiotensin II and norepinephrine caused constriction of the cerebral vessels whereas adenosine caused dilation. Isoproterenol did not affect cerebral arterioles; however, it dilated cheek pouch arterioles. Thus, this preparation provides a satisfactory model for studying the living cerebral microcirculation. V isualization of the surface vessels of the in situ brain has been used extensively to study the cerebral microcirculation. With this technique, it has been shown that topical application of norepinephrine 1 and angiotensin II 2 -3 constricts, whereas adenosine 4 dilates, pial vessels. However, because of the lack of a suitable experimental preparation, no information is available concerning the responsiveness of in vivo cerebral parenchymal vessels to various vasoactive drugs. The present study describes such a preparation in terms of its microvascular morphology and its vascular reactivity to several endogenous agents. The preparation provides a unique model for studying the cerebral microcirculation in normal and pathological states. Materials and Methods Allograft ProcedureNeonatal hamsters (<24 hours old) were used as donors for the cerebral tissue that was subsequently allografted into the cheek pouches of adult female hamsters weighing 120-160 g. This allograft procedure, described by Greenblatt et al 5 and modified by Click et al, 6 was completed under sterile conditions. Using sterile techniques, longitudinal slices of cerebral cortex from decapitated neonate hamsters were placed in sterile lactated Ringer's solution. The recipient hamster was anesthetized with sodium pentobarbital (6.0 mg/100 g body wt i.p.); the hair on the left side of the head was shaved and the skin was cleaned with water, washed with hexachlorophene detergent cleanser, and covered with povidone-iodine. After the cheek pouch was everted and cleaned with warm water, a baseplate attached to a holder was inserted into the cheek pouch. A longitudinal incision was made in the skin exposing the underlying cheek pouch membrane, and the avascular layers were removed. A small slice (<1 x 1 mm square) of cerebral cortex was positioned on the cheek pouch membrane and penicillin G (6,000 units) was placed around the edge of the membrane. The top plate, containing a plastic wrap-covered window, was inserted over the baseplate and sealed. The skin was secured around the top plate with a purse-string suture, and the top plate was sealed with surgical tape. ...