Ganglion cells in the frog retina: discriminant analysis of histological classes.

Neurons in the ganglion cell layer were studied in Golgi-stained flat-mounted frog (Rana temporaria) retinas. Complementary data were obtained from methylene blue- and HRP-stained retinas. On the basis of qualitative criteria, 55 neurons were ordered into six groups, one class of amacrine cell (A1) and five classes of ganglion cells (G1-G5). A discriminant function analysis based on seven morphological variables resulted in a separation of the cell classes in the space of three axes. The A1 cells are small axonless neurons with knotty and dense dendritic trees. The G1 cells are also small, and apparently very numerous, while the G2 cells are medium-sized neurons with two loose dendritic layers, one vitreal and another (less conspicuous) scleral. The rest of the cells are medium-sized to large neurons with sturdy primary dendrites and more distinct dendritic layers, which in some cells (G3) spread both sclerally and vitreally, in other cells in a single either scleral (G4) or vitreal (G5) layer. The relation between our data and the classification of frog ganglion cells recently presented by Frank and Hollyfield is discussed at length, and in that context problems related to statistical classifications are dealt with. A hypothetical identification of the morphological types with the functional cell classes studied in the Helsinki laboratory is discussed.

Comparison of two in vitro methods of bone lead analysis and the implications for in vivo measurements.

Atomic absorption spectrometry and x-ray fluorescence have been used to determine the lead content of metatarsal and tibia bone samples. For a range of bone lead levels from 6.5 to 83 micrograms g-1 of ashed bone there is no evidence of a systematic difference between the two techniques of more than 1 microgram g-1. There is, however, some evidence that random differences between the two in vitro analyses applied to the same bone sample are larger than can be accounted for by known measurement uncertainties. Variations in bone composition could account for these differences. Because the x-ray fluorescence technique is applied in an identical way to in vivo analysis, it is concluded that the uncertainties in in vivo measurements are small.