The world's deep oceans are home to a number of teleosts with asymmetrical or tubular eyes. These immobile eyes possess large spherical lenses and subtend a large binocular visual field directed either dorsally or rostrally. Derived from a lateral non-tubular eye, the tubular eye is comprised of a thick main retina, subserving the rostrally or dorsally directed binocular visual field, and a thin accessory retina subserving, the lateral, monocular visual field. The main retina is thought to receive a focussed image, while the accessory retina is too close to the lens for a focussed image to be received. Several species also possess retinal diverticula, which are small evaginations of differentiated retina located in the rostrolateral wall of the eye and thought to increase the visual field. In order to investigate the spatial resolving power of these retinae (main, accessory and diverticulum), the distribution of cells within the ganglion cell layer was analysed from retinal wholemounts and sectioned material in ten species representing four genera. In all species, the main retina possesses a marked increase in cell density towards a specialised retinal region (area centralis), with a centro-peripheral gradient range between 7:1 and 60:1 and a peak density range of between 30 and 55x10³ cells per mm². The accessory retinae and the transitional zone between the main and accessory retinae possess relatively low cell densities (between 1 and 10 x 10³ cells per mm²) and lack an area centralis. Retinal diverticula examined in four species possess mean ganglion cell densities of between 7.2 and 109.4xl0³ cells per mm². Analyses of soma areas show that the ganglion cell layer of most species possesses cells with areas in a range of 8.0 to 15.4 µm² in the main retina and between 15.1 and 17.4 µm² in the accessory retina. The peak spatial resolving power of the main retina of the ten species varies from 4.1 to 9.1 cycles per degree. The positions of the retinal areae centrales relative to each species' binocular visual field are discussed in relation to what is known of feeding behaviour of these fishes in the deep-sea.

Neonatologists' failure to discuss nonresuscitation options, variations in resuscitation thresholds, and unwillingness to accept nonresuscitation decisions for more mature ELBW infants may restrict parental decision making.

Our findings suggest that maternal cigarette smoking may be the major predictor of tone abnormalities reported in cocaine-exposed infants.

ABSTRACT. Brain glucose concentration during and after hypoxia-ischemia may be one of the variables affecting outcome of asphyxia1 insults. Glucose given before global ischemic forebrain iniurv to adult rats increases morehologic brain damage, i n d postischemic insulin administration reduces selective neuronal necrosis and cortical infarction. Because glucose infusions are routinely used in the clinical management of perinatal asphyxia, we evaluated the role of slucose administration after ischemic neuronal damage to neonatal rat brain. Sprague-Dawley rat pups (postnatal d 7) were subjected to left common carotid artery ligation followed by 2.5 h of 8% oxygen (Levine procedure). The experimental group was subdivided so that pups received either systemic injections of glucose or saline immediately after the hypoxic insult. Animals were killed on postnatal d 12 and brain areas of ipsi-and contralateral cortex and caudate were calculated from camera lucida tracings. There was no significant difference in size of brain infarction between postischemic glucose-treated and postischemic saline-treated pups. However, hypoxic-ischemic brains did show more severe neuronal damage when hyperglycemia was induced after asphyxia. Because postischemic hyperglycemia does not attenuate and may exacerbate injury, we recommend careful monitoring of blood glucose so that hyperglycemia does not occur during resuscitation of asphyxiated infants. (Pediatr Res 32: 489-493, 1992) Abbreviations NADPH-d, NADPH diaphorase EAA, excitatory amino acidNeurologic morbidity occurs in 20-30% of infants suffering acute perinatal asphyxia in the United States (I). The hypoxicischemic encepha~opath~ seen in survivors of pennatal asphyxia is a commonly encountered clinical problem and is thought to be the largest contributor to static encephalopathies in infants and children (2). Hypoglycemia is a frequent metabolic derangement among asphyxiated infants and has additive deleterious effects on the CNS sequelae of perinatal asphyxia (3-5). Standard resuscitation procedures use glucose infusions to correct posthypoxic hypoglycemia, but this procedure often results in hyperglycemia (6).