Two new schemes for fractionation of complex lipid mixtures are presented. Their use for the study of lipids of beef brain, beef liver, and the sea anemone are described. Apparatus and techniques for working in an inert atmosphere, evaporation of solutions in the cold under nitrogen, use of infrared spectroscopy for examination of lipids and their hydrolysis products, preparation and clution of diethylaminoethyl (DEAE) cellulose and silicic acid‐silicate columns and general column combinations that can be used to fractionate complex lipid mixtures are considered in detail. The first scheme, employing DEAE cellulose columns followed by thin layer and paper chromatographic examination of the fractions, was applied to liver lipids. The many components, some of them new lipids not previously detected, are clearly seen with this technique but are not seen when paper or thin layer chromatography alone or silicic acid chromatography are used. The second scheme employing DEAE for initial fractionation, followed by complete separation on silicic acid and silicic acid‐silicate columns, was applied to lipids of the sea anemone and beef brain. Typical lecithin and phosphatidyl ethanolamine were isolated, but sphingomyelin was not found. A new sphingolipid, ceramide aminoethylphosphonate, with a free amino group and a direct carbon to phosphorus bond was isolated and characterized. The methods used for quantitative isolation, the infrared spectra, and the amounts of cholesterol, ceramide, cerebroside, galactosylglyceride, sulfatide, sphingomyelin, lecithin, phosphatidyl ethanolamine, phosphatidyl serine, phosphatidyl inositol, triphosphoinositide, phosphatidic acid, cardiolipin, and ganglioside of beef brain are presented. Finally, the types of lipid‐nonlipid interactions disclosed by column chromatography and their potential application to biological problems are discussed.

No abstract

The homogeneous nucleation of crystals in normal alkane liquids has been studied using a droplet technique for a range of alkanes between C16H34 and C5H12. The critical undercooling for nucleation, ΔTN, relative to the melting point TE, increases significantly with decreasing length of the alkane chain n: ΔTN/TE increases from 0.05 for C16H34 to 0.20 for C5H12, with separate variations noted for the odd-n and even-n alkanes. Use of a spherical nucleus model gives molar surface energies σM relative to the molar heat of fusion, ΔHM, which also increase significantly with decreasing n but show little odd–even variation: σM/ΔHM increases from 0.07 for C16H34 to 0.28 for C5H12. Because of the pronounced increase in surface energy with decreasing chain length, the model of a spherical nucleus is suggested to be inappropriate for the alkane liquids (at least for the high-n alkanes). Using a preferred cylindrical nucleus model with separate side- and end-surface energies, a side-surface energy of about 6 erg cm−2 is estimated from the present data, at least for the high-n alkanes. Using this model, the end-surface energies are found to increase with decreasing chain length. This increase is taken to reflect the increasing importance of chain localization and chain end effects with decreasing chain length. For the small-n alkanes, the side surface energy is also suggested to increase with decreasing chain length, again associated with chain localization effects in the liquid adjacent to the crystal (nucleus) surface.

Improved procedures for preparation of lipid extracts and determination of phospholipids by phosphorus analysis of spots separated by thin layer chromatography (TLC) were employed to determine the phospholipid class distributions of vertebrate (human, bovine, rat, mouse, frog) kidney, liver and spleen. The absence of significant changes arising from postmortem enzymatic degradation was demonstrated by analysis of lipids of organs after standing for different times postmortem. Intraspecies variability was evaluated by separate analysis of several rat organs. Accuracy of analytical results was insured by demonstrating the absence of spot overlap by two‐dimensional TLC and low values for standard deviations. The values for kidney and liver demonstrate little or no species variability, whereas values for spleen indicate two groups which differ in cellular composition. The findings for kidney and liver are in keeping with data obtained from heart, skeletal muscle, lung and highly purified subcellular particulates which indicate that, among vertebrates, there is little or no species variability of phospholipid class distribution of organs and most subcellular particulates.

No abstract